Unnamed: 0.1
int64 0
113k
| Unnamed: 0
float64 0
113k
⌀ | title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|---|---|
106,300 | 106,300 |
Data-Centric Green AI: An Exploratory Empirical Study
|
With the growing availability of large-scale datasets, and the popularization
of affordable storage and computational capabilities, the energy consumed by AI
is becoming a growing concern. To address this issue, in recent years, studies
have focused on demonstrating how AI energy efficiency can be improved by
tuning the model training strategy. Nevertheless, how modifications applied to
datasets can impact the energy consumption of AI is still an open question. To
fill this gap, in this exploratory study, we evaluate if data-centric
approaches can be utilized to improve AI energy efficiency. To achieve our
goal, we conduct an empirical experiment, executed by considering 6 different
AI algorithms, a dataset comprising 5,574 data points, and two dataset
modifications (number of data points and number of features). Our results show
evidence that, by exclusively conducting modifications on datasets, energy
consumption can be drastically reduced (up to 92.16%), often at the cost of a
negligible or even absent accuracy decline. As additional introductory results,
we demonstrate how, by exclusively changing the algorithm used, energy savings
up to two orders of magnitude can be achieved. In conclusion, this exploratory
investigation empirically demonstrates the importance of applying data-centric
techniques to improve AI energy efficiency. Our results call for a research
agenda that focuses on data-centric techniques, to further enable and
democratize Green AI.
|
106,301 | 106,301 |
Walk this Way! Entity Walks and Property Walks for RDF2vec
|
RDF2vec is a knowledge graph embedding mechanism which first extracts
sequences from knowledge graphs by performing random walks, then feeds those
into the word embedding algorithm word2vec for computing vector representations
for entities. In this poster, we introduce two new flavors of walk extraction
coined e-walks and p-walks, which put an emphasis on the structure or the
neighborhood of an entity respectively, and thereby allow for creating
embeddings which focus on similarity or relatedness. By combining the walk
strategies with order-aware and classic RDF2vec, as well as CBOW and skip-gram
word2vec embeddings, we conduct a preliminary evaluation with a total of 12
RDF2vec variants.
|
106,302 | 106,302 |
A Dempster-Shafer approach to trustworthy AI with application to fetal
brain MRI segmentation
|
Deep learning models for medical image segmentation can fail unexpectedly and
spectacularly for pathological cases and for images acquired at different
centers than those used for training, with labeling errors that violate expert
knowledge about the anatomy and the intensity distribution of the regions to be
segmented. Such errors undermine the trustworthiness of deep learning models
developed for medical image segmentation. Mechanisms with a fallback method for
detecting and correcting such failures are essential for safely translating
this technology into clinics and are likely to be a requirement of future
regulations on artificial intelligence (AI). Here, we propose a principled
trustworthy AI theoretical framework and a practical system that can augment
any backbone AI system using a fallback method and a fail-safe mechanism based
on Dempster-Shafer theory. Our approach relies on an actionable definition of
trustworthy AI. Our method automatically discards the voxel-level labeling
predicted by the backbone AI that are likely to violate expert knowledge and
relies on a fallback atlas-based segmentation method for those voxels. We
demonstrate the effectiveness of the proposed trustworthy AI approach on the
largest reported annotated dataset of fetal T2w MRI consisting of 540 manually
annotated fetal brain 3D MRIs with neurotypical or abnormal brain development
and acquired from 13 sources of data across 6 countries. We show that our
trustworthy AI method improves the robustness of a state-of-the-art backbone AI
for fetal brain MRI segmentation on MRIs acquired across various centers and
for fetuses with various brain abnormalities.
|
106,303 | 106,303 |
How Do Graph Networks Generalize to Large and Diverse Molecular Systems?
|
The predominant method of demonstrating progress of atomic graph neural
networks are benchmarks on small and limited datasets. The implicit hypothesis
behind this approach is that progress on these narrow datasets generalize to
the large diversity of chemistry. This generalizability would be very helpful
for research, but currently remains untested. In this work we test this
assumption by identifying four aspects of complexity in which many datasets are
lacking: 1. Chemical diversity (number of different elements), 2. system size
(number of atoms per sample), 3. dataset size (number of data samples), and 4.
domain shift (similarity of the training and test set). We introduce multiple
subsets of the large Open Catalyst 2020 (OC20) dataset to independently
investigate each of these aspects. We then perform 21 ablation studies and
sensitivity analyses on 9 datasets testing both previously proposed and new
model enhancements. We find that some improvements are consistent between
datasets, but many are not and some even have opposite effects. Based on this
analysis, we identify a smaller dataset that correlates well with the full OC20
dataset, and propose the GemNet-OC model, which outperforms the previous
state-of-the-art on OC20 by 16%, while reducing training time by a factor of
10. Overall, our findings challenge the common belief that graph neural
networks work equally well independent of dataset size and diversity, and
suggest that caution must be exercised when making generalizations based on
narrow datasets.
|
106,304 | 106,304 |
Reinforcement Learning Agents in Colonel Blotto
|
Models and games are simplified representations of the world. There are many
different kinds of models, all differing in complexity and which aspect of the
world they allow us to further our understanding of. In this paper we focus on
a specific instance of agent-based models, which uses reinforcement learning
(RL) to train the agent how to act in its environment. Reinforcement learning
agents are usually also Markov processes, which is another type of model that
can be used. We test this reinforcement learning agent in a Colonel Blotto
environment1, and measure its performance against Random agents as its
opponent. We find that the RL agent handily beats a single opponent, and still
performs quite well when the number of opponents are increased. We also analyze
the RL agent and look at what strategies it has arrived by looking at the
actions that it has given the highest and lowest Q-values. Interestingly, the
optimal strategy for playing multiple opponents is almost the complete opposite
of the optimal strategy for playing a single opponent.
|
106,305 | 106,305 |
Classification of NEQR Processed Classical Images using Quantum Neural
Networks (QNN)
|
A quantum neural network (QNN) is interpreted today as any quantum circuit
with trainable continuous parameters. This work builds on previous works by the
authors and addresses QNN for image classification with Novel Enhanced Quantum
Representation of (NEQR) processed classical data where Principal component
analysis (PCA) and Projected Quantum Kernel features (PQK) were investigated
previously by the authors as a path to quantum advantage for the same classical
dataset. For each of these cases the Fashion-MNIST dataset was downscaled using
PCA to convert into quantum data where the classical NN easily outperformed the
QNN. However, we demonstrated quantum advantage by using PQK where quantum
models achieved more than ~90% accuracy surpassing their classical counterpart
on the same training dataset as in the first case. In this current work, we use
the same dataset fed into a QNN and compare that with performance of a
classical NN model. We built an NEQR model circuit to pre-process the same data
and feed the images into the QNN. Our results showed marginal improvements
(only about ~5.0%) where the QNN performance with NEQR exceeded the performance
of QNN without NEQR. We conclude that given the computational cost and the
massive circuit depth associated with running NEQR, the advantage offered by
this specific Quantum Image Processing (QIMP) algorithm is questionable at
least for classical image dataset. No actual quantum computing hardware
platform exists today that can support the circuit depth needed to run NEQR
even for the reduced image sizes of our toy classical dataset.
|
106,306 | 106,306 |
Dimensionality Expansion and Transfer Learning for Next Generation
Energy Management Systems
|
Electrical management systems (EMS) are playing a central role in enabling
energy savings. They can be deployed within an everyday household where they
monitor and manage appliances and help residents be more energy efficient and
subsequently also more economical. One of they key functionalities of EMS is to
automatically detect and identify appliances within a household through the
process of load monitoring. In this paper, we propose a new transfer learning
approach for building EMS (BEMS) and study the trade-offs in terms of numbers
of samples and target classes in adapting a backbone model during the transfer
process. We also perform a first time analysis of feature expansion through
video-like transformation of time series data for device classification in non
intrusive load monitoring (NILM) and propose a deep learning architecture
enabling accurate appliance identification. We examine the relative performance
of our method on 5 different representative low-frequency datasets and show
that our method performs with an average F1 score of 0.88 on these datasets.
|
106,307 | 106,307 |
A Transformer-Based Contrastive Learning Approach for Few-Shot Sign
Language Recognition
|
Sign language recognition from sequences of monocular images or 2D poses is a
challenging field, not only due to the difficulty to infer 3D information from
2D data, but also due to the temporal relationship between the sequences of
information. Additionally, the wide variety of signs and the constant need to
add new ones on production environments makes it infeasible to use traditional
classification techniques. We propose a novel Contrastive Transformer-based
model, which demonstrate to learn rich representations from body key points
sequences, allowing better comparison between vector embedding. This allows us
to apply these techniques to perform one-shot or few-shot tasks, such as
classification and translation. The experiments showed that the model could
generalize well and achieved competitive results for sign classes never seen in
the training process.
|
106,308 | 106,308 |
Federated Self-supervised Speech Representations: Are We There Yet?
|
The ubiquity of microphone-enabled devices has lead to large amounts of
unlabelled audio data being produced at the edge. The integration of
self-supervised learning (SSL) and federated learning (FL) into one coherent
system can potentially offer data privacy guarantees while also advancing the
quality and robustness of speech representations. In this paper, we provide a
first-of-its-kind systematic study of the feasibility and complexities for
training speech SSL models under FL scenarios from the perspective of
algorithms, hardware, and systems limits. Despite the high potential of their
combination, we find existing system constraints and algorithmic behaviour make
SSL and FL systems nearly impossible to build today. Yet critically, our
results indicate specific performance bottlenecks and research opportunities
that would allow this situation to be reversed. While our analysis suggests
that, given existing trends in hardware, hybrid SSL and FL speech systems will
not be viable until 2027. We believe this study can act as a roadmap to
accelerate work towards reaching this milestone much earlier.
|
106,309 | 106,309 |
High Probability Bounds for a Class of Nonconvex Algorithms with AdaGrad
Stepsize
|
In this paper, we propose a new, simplified high probability analysis of
AdaGrad for smooth, non-convex problems. More specifically, we focus on a
particular accelerated gradient (AGD) template (Lan, 2020), through which we
recover the original AdaGrad and its variant with averaging, and prove a
convergence rate of $\mathcal O (1/ \sqrt{T})$ with high probability without
the knowledge of smoothness and variance. We use a particular version of
Freedman's concentration bound for martingale difference sequences (Kakade &
Tewari, 2008) which enables us to achieve the best-known dependence of $\log (1
/ \delta )$ on the probability margin $\delta$. We present our analysis in a
modular way and obtain a complementary $\mathcal O (1 / T)$ convergence rate in
the deterministic setting. To the best of our knowledge, this is the first high
probability result for AdaGrad with a truly adaptive scheme, i.e., completely
oblivious to the knowledge of smoothness and uniform variance bound, which
simultaneously has best-known dependence of $\log( 1/ \delta)$. We further
prove noise adaptation property of AdaGrad under additional noise assumptions.
|
106,310 | 106,310 |
KNN-Diffusion: Image Generation via Large-Scale Retrieval
|
While the availability of massive Text-Image datasets is shown to be
extremely useful in training large-scale generative models (e.g. DDPMs,
Transformers), their output typically depends on the quality of both the input
text, as well as the training dataset. In this work, we show how large-scale
retrieval methods, in particular efficient K-Nearest-Neighbors (KNN) search,
can be used in order to train a model to adapt to new samples. Learning to
adapt enables several new capabilities. Sifting through billions of records at
inference time is extremely efficient and can alleviate the need to train or
memorize an adequately large generative model. Additionally, fine-tuning
trained models to new samples can be achieved by simply adding them to the
table. Rare concepts, even without any presence in the training set, can be
then leveraged during test time without any modification to the generative
model. Our diffusion-based model trains on images only, by leveraging a joint
Text-Image multi-modal metric. Compared to baseline methods, our generations
achieve state of the art results both in human evaluations as well as with
perceptual scores when tested on a public multimodal dataset of natural images,
as well as on a collected dataset of 400 million Stickers.
|
106,311 | 106,311 |
PAnDR: Fast Adaptation to New Environments from Offline Experiences via
Decoupling Policy and Environment Representations
|
Deep Reinforcement Learning (DRL) has been a promising solution to many
complex decision-making problems. Nevertheless, the notorious weakness in
generalization among environments prevent widespread application of DRL agents
in real-world scenarios. Although advances have been made recently, most prior
works assume sufficient online interaction on training environments, which can
be costly in practical cases. To this end, we focus on an
offline-training-online-adaptation setting, in which the agent first learns
from offline experiences collected in environments with different dynamics and
then performs online policy adaptation in environments with new dynamics. In
this paper, we propose Policy Adaptation with Decoupled Representations (PAnDR)
for fast policy adaptation. In offline training phase, the environment
representation and policy representation are learned through contrastive
learning and policy recovery, respectively. The representations are further
refined by mutual information optimization to make them more decoupled and
complete. With learned representations, a Policy-Dynamics Value Function (PDVF)
[Raileanu et al., 2020] network is trained to approximate the values for
different combinations of policies and environments from offline experiences.
In online adaptation phase, with the environment context inferred from few
experiences collected in new environments, the policy is optimized by gradient
ascent with respect to the PDVF. Our experiments show that PAnDR outperforms
existing algorithms in several representative policy adaptation problems.
|
106,312 | 106,312 |
Sub-Task Decomposition Enables Learning in Sequence to Sequence Tasks
|
The field of Natural Language Processing has experienced a dramatic leap in
capabilities with the recent introduction of huge Language Models. Despite this
success, natural language problems that involve several compounded steps are
still practically unlearnable, even by the largest LMs. This complies with
experimental failures for end-to-end learning of composite problems that were
demonstrated in a variety of domains. An effective mitigation is to introduce
intermediate supervision for solving sub-tasks of the compounded problem.
Recently, several works have demonstrated high gains by taking a
straightforward approach for incorporating intermediate supervision in
compounded natural language problems: the sequence-to-sequence LM is fed with
an augmented input, in which the decomposed tasks' labels are simply
concatenated to the original input. In this paper, we prove a positive learning
result that motivates these recent efforts. We show that when concatenating
intermediate supervision to the input and training a sequence-to-sequence model
on this modified input, unlearnable composite problems can become learnable. We
show that this is true for any family of tasks which on the one hand, are
unlearnable, and on the other hand, can be decomposed into a polynomial number
of simple sub-tasks, each of which depends only on O(1) previous sub-task
results. Beyond motivating contemporary empirical efforts for incorporating
intermediate supervision in sequence-to-sequence language models, our positive
theoretical result is the first of its kind in the landscape of results on the
benefits of intermediate supervision for neural-network learning: Until now,
all theoretical results on the subject are negative, i.e., show cases where
learning is impossible without intermediate supervision, while our result is
positive, showing that learning is facilitated in the presence of intermediate
supervision.
|
106,313 | 106,313 |
Efficient Bayesian Network Structure Learning via Parameterized Local
Search on Topological Orderings
|
In Bayesian Network Structure Learning (BNSL), one is given a variable set
and parent scores for each variable and aims to compute a DAG, called Bayesian
network, that maximizes the sum of parent scores, possibly under some
structural constraints. Even very restricted special cases of BNSL are
computationally hard, and, thus, in practice heuristics such as local search
are used. A natural approach for a local search algorithm is a hill climbing
strategy, where one replaces a given BNSL solution by a better solution within
some pre-defined neighborhood as long as this is possible. We study
ordering-based local search, where a solution is described via a topological
ordering of the variables. We show that given such a topological ordering, one
can compute an optimal DAG whose ordering is within inversion distance $r$ in
subexponential FPT time; the parameter $r$ allows to balance between solution
quality and running time of the local search algorithm. This running time bound
can be achieved for BNSL without structural constraints and for all structural
constraints that can be expressed via a sum of weights that are associated with
each parent set. We also introduce a related distance called `window inversions
distance' and show that the corresponding local search problem can also be
solved in subexponential FPT time for the parameter $r$. For two further
natural modification operations on the variable orderings, we show that
algorithms with an FPT time for $r$ are unlikely. We also outline the limits of
ordering-based local search by showing that it cannot be used for common
structural constraints on the moralized graph of the network.
|
106,314 | 106,314 |
Question Generation for Reading Comprehension Assessment by Modeling How
and What to Ask
|
Reading is integral to everyday life, and yet learning to read is a struggle
for many young learners. During lessons, teachers can use comprehension
questions to increase engagement, test reading skills, and improve retention.
Historically such questions were written by skilled teachers, but recently
language models have been used to generate comprehension questions. However,
many existing Question Generation (QG) systems focus on generating literal
questions from the text, and have no way to control the type of the generated
question. In this paper, we study QG for reading comprehension where
inferential questions are critical and extractive techniques cannot be used. We
propose a two-step model (HTA-WTA) that takes advantage of previous datasets,
and can generate questions for a specific targeted comprehension skill. We
propose a new reading comprehension dataset that contains questions annotated
with story-based reading comprehension skills (SBRCS), allowing for a more
complete reader assessment. Across several experiments, our results show that
HTA-WTA outperforms multiple strong baselines on this new dataset. We show that
the HTA-WTA model tests for strong SCRS by asking deep inferential questions.
|
106,315 | 106,315 |
A survey on recently proposed activation functions for Deep Learning
|
Artificial neural networks (ANN), typically referred to as neural networks,
are a class of Machine Learning algorithms and have achieved widespread
success, having been inspired by the biological structure of the human brain.
Neural networks are inherently powerful due to their ability to learn complex
function approximations from data. This generalization ability has been able to
impact multidisciplinary areas involving image recognition, speech recognition,
natural language processing, and others. Activation functions are a crucial
sub-component of neural networks. They define the output of a node in the
network given a set of inputs. This survey discusses the main concepts of
activation functions in neural networks, including; a brief introduction to
deep neural networks, a summary of what are activation functions and how they
are used in neural networks, their most common properties, the different types
of activation functions, some of the challenges, limitations, and alternative
solutions faced by activation functions, concluding with the final remarks.
|
106,316 | 106,316 |
Last Layer Re-Training is Sufficient for Robustness to Spurious
Correlations
|
Neural network classifiers can largely rely on simple spurious features, such
as backgrounds, to make predictions. However, even in these cases, we show that
they still often learn core features associated with the desired attributes of
the data, contrary to recent findings. Inspired by this insight, we demonstrate
that simple last layer retraining can match or outperform state-of-the-art
approaches on spurious correlation benchmarks, but with profoundly lower
complexity and computational expenses. Moreover, we show that last layer
retraining on large ImageNet-trained models can also significantly reduce
reliance on background and texture information, improving robustness to
covariate shift, after only minutes of training on a single GPU.
|
106,317 | 106,317 |
Marrying Fairness and Explainability in Supervised Learning
|
Machine learning algorithms that aid human decision-making may inadvertently
discriminate against certain protected groups. We formalize direct
discrimination as a direct causal effect of the protected attributes on the
decisions, while induced discrimination as a change in the causal influence of
non-protected features associated with the protected attributes. The
measurements of marginal direct effect (MDE) and SHapley Additive exPlanations
(SHAP) reveal that state-of-the-art fair learning methods can induce
discrimination via association or reverse discrimination in synthetic and
real-world datasets. To inhibit discrimination in algorithmic systems, we
propose to nullify the influence of the protected attribute on the output of
the system, while preserving the influence of remaining features. We introduce
and study post-processing methods achieving such objectives, finding that they
yield relatively high model accuracy, prevent direct discrimination, and
diminishes various disparity measures, e.g., demographic disparity.
|
106,318 | 106,318 |
Guaranteed Bounds for Posterior Inference in Universal Probabilistic
Programming
|
We propose a new method to approximate the posterior distribution of
probabilistic programs by means of computing guaranteed bounds. The starting
point of our work is an interval-based trace semantics for a recursive,
higher-order probabilistic programming language with continuous distributions.
Taking the form of (super-/subadditive) measures, these lower/upper bounds are
non-stochastic and provably correct: using the semantics, we prove that the
actual posterior of a given program is sandwiched between the lower and upper
bounds (soundness); moreover the bounds converge to the posterior
(completeness). As a practical and sound approximation, we introduce a
weight-aware interval type system, which automatically infers interval bounds
on not just the return value but also weight of program executions,
simultaneously. We have built a tool implementation, called GuBPI, which
automatically computes these posterior lower/upper bounds. Our evaluation on
examples from the literature shows that the bounds are useful, and can even be
used to recognise wrong outputs from stochastic posterior inference procedures.
|
106,319 | 106,319 |
Simple and Effective Synthesis of Indoor 3D Scenes
|
We study the problem of synthesizing immersive 3D indoor scenes from one or
more images. Our aim is to generate high-resolution images and videos from
novel viewpoints, including viewpoints that extrapolate far beyond the input
images while maintaining 3D consistency. Existing approaches are highly
complex, with many separately trained stages and components. We propose a
simple alternative: an image-to-image GAN that maps directly from reprojections
of incomplete point clouds to full high-resolution RGB-D images. On the
Matterport3D and RealEstate10K datasets, our approach significantly outperforms
prior work when evaluated by humans, as well as on FID scores. Further, we show
that our model is useful for generative data augmentation. A
vision-and-language navigation (VLN) agent trained with trajectories
spatially-perturbed by our model improves success rate by up to 1.5% over a
state of the art baseline on the R2R benchmark. Our code will be made available
to facilitate generative data augmentation and applications to downstream
robotics and embodied AI tasks.
|
106,320 | 106,320 |
LilNetX: Lightweight Networks with EXtreme Model Compression and
Structured Sparsification
|
We introduce LilNetX, an end-to-end trainable technique for neural networks
that enables learning models with specified accuracy-rate-computation
trade-off. Prior works approach these problems one at a time and often require
post-processing or multistage training which become less practical and do not
scale very well for large datasets or architectures. Our method constructs a
joint training objective that penalizes the self-information of network
parameters in a reparameterized latent space to encourage small model size
while also introducing priors to increase structured sparsity in the parameter
space to reduce computation. We achieve up to 50% smaller model size and 98%
model sparsity on ResNet-20 while retaining the same accuracy on the CIFAR-10
dataset as well as 35% smaller model size and 42% structured sparsity on
ResNet-50 trained on ImageNet, when compared to existing state-of-the-art model
compression methods. Code is available at
https://github.com/Sharath-girish/LilNetX.
|
106,321 | 106,321 |
Multi-task nonparallel support vector machine for classification
|
Direct multi-task twin support vector machine (DMTSVM) explores the shared
information between multiple correlated tasks, then it produces better
generalization performance. However, it contains matrix inversion operation
when solving the dual problems, so it costs much running time. Moreover, kernel
trick cannot be directly utilized in the nonlinear case. To effectively avoid
above problems, a novel multi-task nonparallel support vector machine (MTNPSVM)
including linear and nonlinear cases is proposed in this paper. By introducing
epsilon-insensitive loss instead of square loss in DMTSVM, MTNPSVM effectively
avoids matrix inversion operation and takes full advantage of the kernel trick.
Theoretical implication of the model is further discussed. To further improve
the computational efficiency, the alternating direction method of multipliers
(ADMM) is employed when solving the dual problem. The computational complexity
and convergence of the algorithm are provided. In addition, the property and
sensitivity of the parameter in model are further explored. The experimental
results on fifteen benchmark datasets and twelve image datasets demonstrate the
validity of MTNPSVM in comparison with the state-of-the-art algorithms.
Finally, it is applied to real Chinese Wine dataset, and also verifies its
effectiveness.
|
106,322 | 106,322 |
Incremental Unsupervised Feature Selection for Dynamic Incomplete
Multi-view Data
|
Multi-view unsupervised feature selection has been proven to be efficient in
reducing the dimensionality of multi-view unlabeled data with high dimensions.
The previous methods assume all of the views are complete. However, in real
applications, the multi-view data are often incomplete, i.e., some views of
instances are missing, which will result in the failure of these methods.
Besides, while the data arrive in form of streams, these existing methods will
suffer the issues of high storage cost and expensive computation time. To
address these issues, we propose an Incremental Incomplete Multi-view
Unsupervised Feature Selection method (I$^2$MUFS) on incomplete multi-view
streaming data. By jointly considering the consistent and complementary
information across different views, I$^2$MUFS embeds the unsupervised feature
selection into an extended weighted non-negative matrix factorization model,
which can learn a consensus clustering indicator matrix and fuse different
latent feature matrices with adaptive view weights. Furthermore, we introduce
the incremental leaning mechanisms to develop an alternative iterative
algorithm, where the feature selection matrix is incrementally updated, rather
than recomputing on the entire updated data from scratch. A series of
experiments are conducted to verify the effectiveness of the proposed method by
comparing with several state-of-the-art methods. The experimental results
demonstrate the effectiveness and efficiency of the proposed method in terms of
the clustering metrics and the computational cost.
|
106,323 | 106,323 |
End-To-End Optimization of Online Neural Network-supported Two-Stage
Dereverberation for Hearing Devices
|
A two-stage online dereverberation algorithm for hearing devices is presented
in this paper. The approach combines a multi-channel multi-frame linear
filtering approach with a single-channel single-frame post-filter. Both
components rely on power spectral density (PSD) estimates provided by deep
neural networks (DNNs). This contribution extends our prior work, which shows
that directly optimizing for a criterion at the output of the multi-channel
linear filtering stage results in a more efficient dereverberation, as compared
to placing the criterion at the output of the DNN to optimize the PSD
estimation. In the present work, we show that the dereverberation performance
of the proposed first stage particularly improves the early-to-mid
reverberation ratio if trained end-to-end. We thus argue that it can be
combined with a post-filtering stage which benefits from the early-to-mid ratio
improvement and is consequently able to efficiently suppress the residual late
reverberation. This proposed two stage procedure is shown to be both very
effective in terms of dereverberation performance and computational demands.
Furthermore, the proposed system can be adapted to the needs of different types
of hearing-device users by controlling the amount of reduction of early
reflections. The proposed system outperforms the previously proposed end-to-end
DNN-supported linear filtering algorithm, as well as other traditional
approaches, based on an evaluation using the noise-free version of the WHAMR!
dataset.
|
106,324 | 106,324 |
Federated Learning for Distributed Spectrum Sensing in NextG
Communication Networks
|
NextG networks are intended to provide the flexibility of sharing the
spectrum with incumbent users and support various spectrum monitoring tasks
such as anomaly detection, fault diagnostics, user equipment identification,
and authentication. A network of wireless sensors is needed to monitor the
spectrum for signal transmissions of interest over a large deployment area.
Each sensor receives signals under a specific channel condition depending on
its location and trains an individual model of a deep neural network (DNN)
accordingly to classify signals. To improve the accuracy, individual sensors
may exchange sensing data or sensor results with each other or with a fusion
center (such as in cooperative spectrum sensing). In this paper, distributed
federated learning over a multi-hop wireless network is considered to
collectively train a DNN for signal identification. In distributed federated
learning, each sensor broadcasts its trained model to its neighbors, collects
the DNN models from its neighbors, and aggregates them to initialize its own
model for the next round of training. Without exchanging any spectrum data,
this process is repeated over time such that a common DNN is built across the
network while preserving the privacy associated with signals collected at
different locations. Signal classification accuracy and convergence time are
evaluated for different network topologies (including line, star, ring, grid,
and random networks) and packet loss events. Then, the reduction of
communication overhead and energy consumption is considered with random
participation of sensors in model updates. The results show the feasibility of
extending cooperative spectrum sensing over a general multi-hop wireless
network through federated learning and indicate its robustness to wireless
network effects, thereby sustaining high accuracy with low communication
overhead and energy consumption.
|
106,325 | 106,325 |
Statistical Model Criticism of Variational Auto-Encoders
|
We propose a framework for the statistical evaluation of variational
auto-encoders (VAEs) and test two instances of this framework in the context of
modelling images of handwritten digits and a corpus of English text. Our take
on evaluation is based on the idea of statistical model criticism, popular in
Bayesian data analysis, whereby a statistical model is evaluated in terms of
its ability to reproduce statistics of an unknown data generating process from
which we can obtain samples. A VAE learns not one, but two joint distributions
over a shared sample space, each exploiting a choice of factorisation that
makes sampling tractable in one of two directions (latent-to-data,
data-to-latent). We evaluate samples from these distributions, assessing their
(marginal) fit to the observed data and our choice of prior, and we also
evaluate samples through a pipeline that connects the two distributions
starting from a data sample, assessing whether together they exploit and reveal
latent factors of variation that are useful to a practitioner. We show that
this methodology offers possibilities for model selection qualitatively beyond
intrinsic evaluation metrics and at a finer granularity than commonly used
statistics can offer.
|
106,326 | 106,326 |
SOMOS: The Samsung Open MOS Dataset for the Evaluation of Neural
Text-to-Speech Synthesis
|
In this work, we present the SOMOS dataset, the first large-scale mean
opinion scores (MOS) dataset consisting of solely neural text-to-speech (TTS)
samples. It can be employed to train automatic MOS prediction systems focused
on the assessment of modern synthesizers, and can stimulate advancements in
acoustic model evaluation. It consists of 20K synthetic utterances of the LJ
Speech voice, a public domain speech dataset which is a common benchmark for
building neural acoustic models and vocoders. Utterances are generated from 200
TTS systems including vanilla neural acoustic models as well as models which
allow prosodic variations. An LPCNet vocoder is used for all systems, so that
the samples' variation depends only on the acoustic models. The synthesized
utterances provide balanced and adequate domain and length coverage. We collect
MOS naturalness evaluations on 3 English Amazon Mechanical Turk locales and
share practices leading to reliable crowdsourced annotations for this task.
Baseline results of state-of-the-art MOS prediction models on the SOMOS dataset
are presented, while we show the challenges that such models face when assigned
to evaluate synthetic utterances.
|
106,327 | 106,327 |
Fusing finetuned models for better pretraining
|
Pretrained models are the standard starting point for training. This approach
consistently outperforms the use of a random initialization. However,
pretraining is a costly endeavour that few can undertake.
In this paper, we create better base models at hardly any cost, by fusing
multiple existing fine tuned models into one. Specifically, we fuse by
averaging the weights of these models. We show that the fused model results
surpass the pretrained model ones. We also show that fusing is often better
than intertraining.
We find that fusing is less dependent on the target task. Furthermore, weight
decay nullifies intertraining effects but not those of fusing.
|
106,328 | 106,328 |
Perceive, Represent, Generate: Translating Multimodal Information to
Robotic Motion Trajectories
|
We present Perceive-Represent-Generate (PRG), a novel three-stage framework
that maps perceptual information of different modalities (e.g., visual or
sound), corresponding to a sequence of instructions, to an adequate sequence of
movements to be executed by a robot. In the first stage, we perceive and
pre-process the given inputs, isolating individual commands from the complete
instruction provided by a human user. In the second stage we encode the
individual commands into a multimodal latent space, employing a deep generative
model. Finally, in the third stage we convert the multimodal latent values into
individual trajectories and combine them into a single dynamic movement
primitive, allowing its execution in a robotic platform. We evaluate our
pipeline in the context of a novel robotic handwriting task, where the robot
receives as input a word through different perceptual modalities (e.g., image,
sound), and generates the corresponding motion trajectory to write it, creating
coherent and readable handwritten words.
|
106,329 | 106,329 |
Standardized feature extraction from pairwise conflicts applied to the
train rescheduling problem
|
We propose a train rescheduling algorithm which applies a standardized
feature selection based on pairwise conflicts in order to serve as input for
the reinforcement learning framework. We implement an analytical method which
identifies and optimally solves every conflict arising between two trains, then
we design a corresponding observation space which features the most relevant
information considering these conflicts. The data obtained this way then
translates to actions in the context of the reinforcement learning framework.
We test our preliminary model using the evaluation metrics of the Flatland
Challenge. The empirical results indicate that the suggested feature space
provides meaningful observations, from which a sensible scheduling policy can
be learned.
|
106,330 | 106,330 |
Graph Neural Networks Designed for Different Graph Types: A Survey
|
Graphs are ubiquitous in nature and can therefore serve as models for many
practical but also theoretical problems. Based on this, the young research
field of Graph Neural Networks (GNNs) has emerged. Despite the youth of the
field and the speed in which new models are developed, many good surveys have
been published in the last years. Nevertheless, an overview on which graph
types can be modeled by GNNs is missing. In this survey, we give a detailed
overview of already existing GNNs and, unlike previous surveys, categorize them
according to their ability to handle different graph types and properties. We
consider GNNs operating on static as well as on dynamic graphs of different
structural constitutions, with or without node or edge attributes. Moreover in
the dynamic case, we separate the models in discrete-time and continuous-time
dynamic graphs based on their architecture. While ordering the existing GNN
models, we find, that there are still graph types, that are not or only rarely
covered by existing GNN models. We point out where models are missing and give
potential reasons for their absence.
|
106,331 | 106,331 |
Knowledge Infused Decoding
|
Pre-trained language models (LMs) have been shown to memorize a substantial
amount of knowledge from the pre-training corpora; however, they are still
limited in recalling factually correct knowledge given a certain context.
Hence, they tend to suffer from counterfactual or hallucinatory generation when
used in knowledge-intensive natural language generation (NLG) tasks. Recent
remedies to this problem focus on modifying either the pre-training or task
fine-tuning objectives to incorporate knowledge, which normally require
additional costly training or architecture modification of LMs for practical
applications. We present Knowledge Infused Decoding (KID) -- a novel decoding
algorithm for generative LMs, which dynamically infuses external knowledge into
each step of the LM decoding. Specifically, we maintain a local knowledge
memory based on the current context, interacting with a dynamically created
external knowledge trie, and continuously update the local memory as a
knowledge-aware constraint to guide decoding via reinforcement learning. On six
diverse knowledge-intensive NLG tasks, task-agnostic LMs (e.g., GPT-2 and BART)
armed with KID outperform many task-optimized state-of-the-art models, and show
particularly strong performance in few-shot scenarios over seven related
knowledge-infusion techniques. Human evaluation confirms KID's ability to
generate more relevant and factual language for the input context when compared
with multiple baselines. Finally, KID also alleviates exposure bias and
provides stable generation quality when generating longer sequences. Code for
KID is available at https://github.com/microsoft/KID.
|
106,332 | 106,332 |
Data Justice Stories: A Repository of Case Studies
|
The idea of "data justice" is of recent academic vintage. It has arisen over
the past decade in Anglo-European research institutions as an attempt to bring
together a critique of the power dynamics that underlie accelerating trends of
datafication with a normative commitment to the principles of social justice-a
commitment to the achievement of a society that is equitable, fair, and capable
of confronting the root causes of injustice.However, despite the seeming
novelty of such a data justice pedigree, this joining up of the critique of the
power imbalances that have shaped the digital and "big data" revolutions with a
commitment to social equity and constructive societal transformation has a
deeper historical, and more geographically diverse, provenance. As the stories
of the data justice initiatives, activism, and advocacy contained in this
volume well evidence, practices of data justice across the globe have, in fact,
largely preceded the elaboration and crystallisation of the idea of data
justice in contemporary academic discourse. In telling these data justice
stories, we hope to provide the reader with two interdependent tools of data
justice thinking: First, we aim to provide the reader with the critical
leverage needed to discern those distortions and malformations of data justice
that manifest in subtle and explicit forms of power, domination, and coercion.
Second, we aim to provide the reader with access to the historically effective
forms of normativity and ethical insight that have been marshalled by data
justice activists and advocates as tools of societal transformation-so that
these forms of normativity and insight can be drawn on, in turn, as
constructive resources to spur future transformative data justice practices.
|
106,333 | 106,333 |
AUV-Net: Learning Aligned UV Maps for Texture Transfer and Synthesis
|
In this paper, we address the problem of texture representation for 3D shapes
for the challenging and underexplored tasks of texture transfer and synthesis.
Previous works either apply spherical texture maps which may lead to large
distortions, or use continuous texture fields that yield smooth outputs lacking
details. We argue that the traditional way of representing textures with images
and linking them to a 3D mesh via UV mapping is more desirable, since
synthesizing 2D images is a well-studied problem. We propose AUV-Net which
learns to embed 3D surfaces into a 2D aligned UV space, by mapping the
corresponding semantic parts of different 3D shapes to the same location in the
UV space. As a result, textures are aligned across objects, and can thus be
easily synthesized by generative models of images. Texture alignment is learned
in an unsupervised manner by a simple yet effective texture alignment module,
taking inspiration from traditional works on linear subspace learning. The
learned UV mapping and aligned texture representations enable a variety of
applications including texture transfer, texture synthesis, and textured single
view 3D reconstruction. We conduct experiments on multiple datasets to
demonstrate the effectiveness of our method. Project page:
https://nv-tlabs.github.io/AUV-NET.
|
106,334 | 106,334 |
Deep transfer learning for system identification using long short-term
memory neural networks
|
Recurrent neural networks (RNNs) have many advantages over more traditional
system identification techniques. They may be applied to linear and nonlinear
systems, and they require fewer modeling assumptions. However, these neural
network models may also need larger amounts of data to learn and generalize.
Furthermore, neural networks training is a time-consuming process. Hence,
building upon long-short term memory neural networks (LSTM), this paper
proposes using two types of deep transfer learning, namely parameter
fine-tuning and freezing, to reduce the data and computation requirements for
system identification. We apply these techniques to identify two dynamical
systems, namely a second-order linear system and a Wiener-Hammerstein nonlinear
system. Results show that compared with direct learning, our method accelerates
learning by 10% to 50%, which also saves data and computing resources.
|
106,335 | 106,335 |
First-Order Algorithms for Nonlinear Generalized Nash Equilibrium
Problems
|
We consider the problem of computing an equilibrium in a class of nonlinear
generalized Nash equilibrium problems (NGNEPs) in which the strategy sets for
each player are defined by equality and inequality constraints that may depend
on the choices of rival players. While the asymptotic global convergence and
local convergence rate of solution procedures have been studied in this
setting, the analysis of iteration complexity is still in its infancy. Our
contribution is to provide two simple first-order algorithmic frameworks based
on the quadratic penalty method and the augmented Lagrangian method,
respectively, with an accelerated mirror-prox algorithm as the inner loop. We
provide nonasymptotic theoretical guarantees for these algorithms. More
specifically, we establish the global convergence rate of our algorithms for
solving (strongly) monotone NGNEPs and we provide iteration complexity bounds
expressed in terms of the number of gradient evaluations. Experimental results
demonstrate the efficiency of our algorithms.
|
106,336 | 106,336 |
DiffCloud: Real-to-Sim from Point Clouds with Differentiable Simulation
and Rendering of Deformable Objects
|
Research in manipulation of deformable objects is typically conducted on a
limited range of scenarios, because handling each scenario on hardware takes
significant effort. Realistic simulators with support for various types of
deformations and interactions have the potential to speed up experimentation
with novel tasks and algorithms. However, for highly deformable objects it is
challenging to align the output of a simulator with the behavior of real
objects. Manual tuning is not intuitive, hence automated methods are needed. We
view this alignment problem as a joint perception-inference challenge and
demonstrate how to use recent neural network architectures to successfully
perform simulation parameter inference from real point clouds. We analyze the
performance of various architectures, comparing their data and training
requirements. Furthermore, we propose to leverage differentiable point cloud
sampling and differentiable simulation to significantly reduce the time to
achieve the alignment. We employ an efficient way to propagate gradients from
point clouds to simulated meshes and further through to the physical simulation
parameters, such as mass and stiffness. Experiments with highly deformable
objects show that our method can achieve comparable or better alignment with
real object behavior, while reducing the time needed to achieve this by more
than an order of magnitude. Videos and supplementary material are available at
https://tinyurl.com/diffcloud.
|
106,337 | 106,337 |
Learning and Transferring Value Function for Robot Exploration in
Subterranean Environments
|
In traditional robot exploration methods, the robot usually does not have
prior biases about the environment it is exploring. Thus the robot assigns
equal importance to the goals which leads to insufficient exploration
efficiency. Alternative, often a hand-tuned policy is used to tweak the value
of goals. In this paper, we present a method to learn how "good" some states
are, measured by the state value function, to provide a hint for the robot to
make exploration decisions. We propose to learn state value functions from
previous offline collected datasets and then transfer and improve the value
function during testing in a new environment. Moreover, the environments
usually have very few and even no extrinsic reward or feedback for the robot.
Therefore in this work, we also tackle the problem of sparse extrinsic rewards
from the environments. We design several intrinsic rewards to encourage the
robot to obtain more information during exploration. These reward functions
then become the building blocks of the state value functions. We test our
method on challenging subterranean and urban environments. To the best of our
knowledge, this work for the first time demonstrates value function prediction
with previous collected datasets to help exploration in challenging
subterranean environments.
|
106,338 | 106,338 |
DeepTensor: Low-Rank Tensor Decomposition with Deep Network Priors
|
DeepTensor is a computationally efficient framework for low-rank
decomposition of matrices and tensors using deep generative networks. We
decompose a tensor as the product of low-rank tensor factors (e.g., a matrix as
the outer product of two vectors), where each low-rank tensor is generated by a
deep network (DN) that is trained in a self-supervised manner to minimize the
mean-squared approximation error. Our key observation is that the implicit
regularization inherent in DNs enables them to capture nonlinear signal
structures (e.g., manifolds) that are out of the reach of classical linear
methods like the singular value decomposition (SVD) and principal component
analysis (PCA). Furthermore, in contrast to the SVD and PCA, whose performance
deteriorates when the tensor's entries deviate from additive white Gaussian
noise, we demonstrate that the performance of DeepTensor is robust to a wide
range of distributions. We validate that DeepTensor is a robust and
computationally efficient drop-in replacement for the SVD, PCA, nonnegative
matrix factorization (NMF), and similar decompositions by exploring a range of
real-world applications, including hyperspectral image denoising, 3D MRI
tomography, and image classification. In particular, DeepTensor offers a 6dB
signal-to-noise ratio improvement over standard denoising methods for signals
corrupted by Poisson noise and learns to decompose 3D tensors 60 times faster
than a single DN equipped with 3D convolutions.
|
106,339 | 106,339 |
Optimization Models and Interpretations for Three Types of Adversarial
Perturbations against Support Vector Machines
|
Adversarial perturbations have drawn great attentions in various deep neural
networks. Most of them are computed by iterations and cannot be interpreted
very well. In contrast, little attentions are paid to basic machine learning
models such as support vector machines. In this paper, we investigate the
optimization models and the interpretations for three types of adversarial
perturbations against support vector machines, including sample-adversarial
perturbations (sAP), class-universal adversarial perturbations (cuAP) as well
as universal adversarial perturbations (uAP). For linear binary/multi
classification support vector machines (SVMs), we derive the explicit solutions
for sAP, cuAP and uAP (binary case), and approximate solution for uAP of
multi-classification. We also obtain the upper bound of fooling rate for uAP.
Such results not only increase the interpretability of the three adversarial
perturbations, but also provide great convenience in computation since
iterative process can be avoided. Numerical results show that our method is
fast and effective in calculating three types of adversarial perturbations.
|
106,340 | 106,340 |
Enhancement on Model Interpretability and Sleep Stage Scoring
Performance with A Novel Pipeline Based on Deep Neural Network
|
Considering the natural frequency characteristics in sleep medicine, this
paper first proposes a time-frequency framework for the representation learning
of the electroencephalogram (EEG) following the definition of the American
Academy of Sleep Medicine. To meet the temporal-random and transient nature of
the defining characteristics of sleep stages, we further design a
context-sensitive flexible pipeline that automatically adapts to the attributes
of data itself. That is, the input EEG spectrogram is partitioned into a
sequence of patches in the time and frequency axes, and then input to a
delicate deep learning network for further representation learning to extract
the stage-dependent features, which are used in the classification step
finally. The proposed pipeline is validated against a large database, i.e., the
Sleep Heart Health Study (SHHS), and the results demonstrate that the
competitive performance for the wake, N2, and N3 stages outperforms the
state-of-art works, with the F1 scores being 0.93, 0.88, and 0.87,
respectively, and the proposed method has a high inter-rater reliability of
0.80 kappa. Importantly, we visualize the stage scoring process of the model
decision with the Layer-wise Relevance Propagation (LRP) method, which shows
that the proposed pipeline is more sensitive and perceivable in the
decision-making process than the baseline pipelines. Therefore, the pipeline
together with the LRP method can provide better model interpretability, which
is important for clinical support.
|
106,341 | 106,341 |
FedCos: A Scene-adaptive Federated Optimization Enhancement for
Performance Improvement
|
As an emerging technology, federated learning (FL) involves training machine
learning models over distributed edge devices, which attracts sustained
attention and has been extensively studied. However, the heterogeneity of
client data severely degrades the performance of FL compared with that in
centralized training. It causes the locally trained models of clients to move
in different directions. On the one hand, it slows down or even stalls the
global updates, leading to inefficient communication. On the other hand, it
enlarges the distances between local models, resulting in an aggregated global
model with poor performance. Fortunately, these shortcomings can be mitigated
by reducing the angle between the directions that local models move in. Based
on this fact, we propose FedCos, which reduces the directional inconsistency of
local models by introducing a cosine-similarity penalty. It promotes the local
model iterations towards an auxiliary global direction. Moreover, our approach
is auto-adapt to various non-IID settings without an elaborate selection of
hyperparameters. The experimental results show that FedCos outperforms the
well-known baselines and can enhance them under a variety of FL scenes,
including varying degrees of data heterogeneity, different number of
participants, and cross-silo and cross-device settings. Besides, FedCos
improves communication efficiency by 2 to 5 times. With the help of FedCos,
multiple FL methods require significantly fewer communication rounds than
before to obtain a model with comparable performance.
|
106,342 | 106,342 |
Distributed Statistical Min-Max Learning in the Presence of Byzantine
Agents
|
Recent years have witnessed a growing interest in the topic of min-max
optimization, owing to its relevance in the context of generative adversarial
networks (GANs), robust control and optimization, and reinforcement learning.
Motivated by this line of work, we consider a multi-agent min-max learning
problem, and focus on the emerging challenge of contending with worst-case
Byzantine adversarial agents in such a setup. By drawing on recent results from
robust statistics, we design a robust distributed variant of the extra-gradient
algorithm - a popular algorithmic approach for min-max optimization. Our main
contribution is to provide a crisp analysis of the proposed robust
extra-gradient algorithm for smooth convex-concave and smooth strongly
convex-strongly concave functions. Specifically, we establish statistical rates
of convergence to approximate saddle points. Our rates are near-optimal, and
reveal both the effect of adversarial corruption and the benefit of
collaboration among the non-faulty agents. Notably, this is the first paper to
provide formal theoretical guarantees for large-scale distributed min-max
learning in the presence of adversarial agents.
|
106,343 | 106,343 |
MultiAuto-DeepONet: A Multi-resolution Autoencoder DeepONet for
Nonlinear Dimension Reduction, Uncertainty Quantification and Operator
Learning of Forward and Inverse Stochastic Problems
|
A new data-driven method for operator learning of stochastic differential
equations(SDE) is proposed in this paper. The central goal is to solve forward
and inverse stochastic problems more effectively using limited data. Deep
operator network(DeepONet) has been proposed recently for operator learning.
Compared to other neural networks to learn functions, it aims at the problem of
learning nonlinear operators. However, it can be challenging by using the
original model to learn nonlinear operators for high-dimensional stochastic
problems. We propose a new multi-resolution autoencoder DeepONet model referred
to as MultiAuto-DeepONet to deal with this difficulty with the aid of
convolutional autoencoder. The encoder part of the network is designed to
reduce the dimensionality as well as discover the hidden features of
high-dimensional stochastic inputs. The decoder is designed to have a special
structure, i.e. in the form of DeepONet. The first DeepONet in decoder is
designed to reconstruct the input function involving randomness while the
second one is used to approximate the solution of desired equations. Those two
DeepONets has a common branch net and two independent trunk nets. This
architecture enables us to deal with multi-resolution inputs naturally. By
adding $L_1$ regularization to our network, we found the outputs from the
branch net and two trunk nets all have sparse structures. This reduces the
number of trainable parameters in the neural network thus making the model more
efficient. Finally, we conduct several numerical experiments to illustrate the
effectiveness of our proposed MultiAuto-DeepONet model with uncertainty
quantification.
|
106,344 | 106,344 |
A Joint Learning Approach for Semi-supervised Neural Topic Modeling
|
Topic models are some of the most popular ways to represent textual data in
an interpret-able manner. Recently, advances in deep generative models,
specifically auto-encoding variational Bayes (AEVB), have led to the
introduction of unsupervised neural topic models, which leverage deep
generative models as opposed to traditional statistics-based topic models. We
extend upon these neural topic models by introducing the Label-Indexed Neural
Topic Model (LI-NTM), which is, to the extent of our knowledge, the first
effective upstream semi-supervised neural topic model. We find that LI-NTM
outperforms existing neural topic models in document reconstruction benchmarks,
with the most notable results in low labeled data regimes and for data-sets
with informative labels; furthermore, our jointly learned classifier
outperforms baseline classifiers in ablation studies.
|
106,345 | 106,345 |
Transformer-Based Language Models for Software Vulnerability Detection:
Performance, Model's Security and Platforms
|
The large transformer-based language models demonstrate excellent performance
in natural language processing. By considering the closeness of natural
languages to the high-level programming language such as C/C++, this work
studies how good are the large transformer-based language models detecting
software vulnerabilities. Our results demonstrate the well performance of these
models on software vulnerability detection. The answer enables extending
transformer-based language models to vulnerability detection and leveraging
superior performance beyond the natural language processing domain. Besides, we
perform the model's security check using Microsoft's Counterfit, a command-line
tool to assess the model's security. Our results find that these models are
vulnerable to adversarial examples. In this regard, we present a simple
countermeasure and its result. Experimenting with large models is always a
challenge due to the requirement of computing resources and platforms/libraries
& dependencies. Based on the experiences and difficulties we faced during this
work, we present our recommendation while choosing the platforms to run these
large models. Moreover, the popular platforms are surveyed thoroughly in this
paper.
|
106,346 | 106,346 |
Neural Implicit Flow: a mesh-agnostic dimensionality reduction paradigm
of spatio-temporal data
|
High-dimensional spatio-temporal dynamics can often be encoded in a
low-dimensional subspace. Engineering applications for modeling,
characterization, design, and control of such large-scale systems often rely on
dimensionality reduction to make solutions computationally tractable in
real-time. Common existing paradigms for dimensionality reduction include
linear methods, such as the singular value decomposition (SVD), and nonlinear
methods, such as variants of convolutional autoencoders (CAE). However, these
encoding techniques lack the ability to efficiently represent the complexity
associated with spatio-temporal data, which often requires variable geometry,
non-uniform grid resolution, adaptive meshing, and/or parametric dependencies.
To resolve these practical engineering challenges, we propose a general
framework called Neural Implicit Flow (NIF) that enables a mesh-agnostic,
low-rank representation of large-scale, parametric, spatial-temporal data. NIF
consists of two modified multilayer perceptrons (MLPs): (i) ShapeNet, which
isolates and represents the spatial complexity, and (ii) ParameterNet, which
accounts for any other input complexity, including parametric dependencies,
time, and sensor measurements. We demonstrate the utility of NIF for parametric
surrogate modeling, enabling the interpretable representation and compression
of complex spatio-temporal dynamics, efficient many-spatial-query tasks, and
improved generalization performance for sparse reconstruction.
|
106,347 | 106,347 |
DDOS: A MOS Prediction Framework utilizing Domain Adaptive Pre-training
and Distribution of Opinion Scores
|
Mean opinion score (MOS) is a typical subjective evaluation metric for speech
synthesis systems. Since collecting MOS is time-consuming, it would be
desirable if there are accurate MOS prediction models for automatic evaluation.
In this work, we propose DDOS, a novel MOS prediction model. DDOS utilizes
domain adaptive pre-training to further pre-train self-supervised learning
models on synthetic speech. And a proposed module is added to model the opinion
score distribution of each utterance. With the proposed components, DDOS
outperforms previous works on BVCC dataset. And the zero shot transfer result
on BC2019 dataset is significantly improved. DDOS also wins second place in
Interspeech 2022 VoiceMOS challenge in terms of system-level score.
|
106,348 | 106,348 |
Explicit Feature Interaction-aware Graph Neural Networks
|
Graph neural networks are powerful methods to handle graph-structured data.
However, existing graph neural networks only learn higher-order feature
interactions implicitly. Thus, they cannot capture information that occurred in
low-order feature interactions. To overcome this problem, we propose Explicit
Feature Interaction-aware Graph Neural Network (EFI-GNN), which explicitly
learns arbitrary-order feature interactions. EFI-GNN can jointly learn with any
other graph neural network. We demonstrate that the joint learning method
always enhances performance on the various node classification tasks.
Furthermore, since EFI-GNN is inherently a linear model, we can interpret the
prediction result of EFI-GNN. With the computation rule, we can obtain an
any-order feature's effect on the decision. By that, we visualize the effects
of the first-order and second-order features as a form of a heatmap.
|
106,349 | 106,349 |
Accelerating Attention through Gradient-Based Learned Runtime Pruning
|
Self-attention is a key enabler of state-of-art accuracy for various
transformer-based Natural Language Processing models. This attention mechanism
calculates a correlation score for each word with respect to the other words in
a sentence. Commonly, only a small subset of words highly correlates with the
word under attention, which is only determined at runtime. As such, a
significant amount of computation is inconsequential due to low attention
scores and can potentially be pruned. The main challenge is finding the
threshold for the scores below which subsequent computation will be
inconsequential. Although such a threshold is discrete, this paper formulates
its search through a soft differentiable regularizer integrated into the loss
function of the training. This formulation piggy backs on the back-propagation
training to analytically co-optimize the threshold and the weights
simultaneously, striking a formally optimal balance between accuracy and
computation pruning. To best utilize this mathematical innovation, we devise a
bit-serial architecture, dubbed LeOPArd, for transformer language models with
bit-level early termination microarchitectural mechanism. We evaluate our
design across 43 back-end tasks for MemN2N, BERT, ALBERT, GPT-2, and Vision
transformer models. Post-layout results show that, on average, LeOPArd yields
1.9x and 3.9x speedup and energy reduction, respectively, while keeping the
average accuracy virtually intact (<0.2% degradation)
|
106,350 | 106,350 |
What You See is What You Get: Distributional Generalization for
Algorithm Design in Deep Learning
|
We investigate and leverage a connection between Differential Privacy (DP)
and the recently proposed notion of Distributional Generalization (DG).
Applying this connection, we introduce new conceptual tools for designing
deep-learning methods that bypass "pathologies" of standard stochastic gradient
descent (SGD). First, we prove that differentially private methods satisfy a
"What You See Is What You Get (WYSIWYG)" generalization guarantee: whatever a
model does on its train data is almost exactly what it will do at test time.
This guarantee is formally captured by distributional generalization. WYSIWYG
enables principled algorithm design in deep learning by reducing
$\textit{generalization}$ concerns to $\textit{optimization}$ ones: in order to
mitigate unwanted behavior at test time, it is provably sufficient to mitigate
this behavior on the train data. This is notably false for standard (non-DP)
methods, hence this observation has applications even when privacy is not
required. For example, importance sampling is known to fail for standard SGD,
but we show that it has exactly the intended effect for DP-trained models.
Thus, with DP-SGD, unlike with SGD, we can influence test-time behavior by
making principled train-time interventions. We use these insights to construct
simple algorithms which match or outperform SOTA in several distributional
robustness applications, and to significantly improve the privacy vs. disparate
impact trade-off of DP-SGD. Finally, we also improve on known theoretical
bounds relating differential privacy, stability, and distributional
generalization.
|
106,351 | 106,351 |
Learning to Solve Travelling Salesman Problem with Hardness-adaptive
Curriculum
|
Various neural network models have been proposed to tackle combinatorial
optimization problems such as the travelling salesman problem (TSP). Existing
learning-based TSP methods adopt a simple setting that the training and testing
data are independent and identically distributed. However, the existing
literature fails to solve TSP instances when training and testing data have
different distributions. Concretely, we find that different training and
testing distribution will result in more difficult TSP instances, i.e., the
solution obtained by the model has a large gap from the optimal solution. To
tackle this problem, in this work, we study learning-based TSP methods when
training and testing data have different distributions using adaptive-hardness,
i.e., how difficult a TSP instance can be for a solver. This problem is
challenging because it is non-trivial to (1) define hardness measurement
quantitatively; (2) efficiently and continuously generate sufficiently hard TSP
instances upon model training; (3) fully utilize instances with different
levels of hardness to learn a more powerful TSP solver. To solve these
challenges, we first propose a principled hardness measurement to quantify the
hardness of TSP instances. Then, we propose a hardness-adaptive generator to
generate instances with different hardness. We further propose a curriculum
learner fully utilizing these instances to train the TSP solver. Experiments
show that our hardness-adaptive generator can generate instances ten times
harder than the existing methods, and our proposed method achieves significant
improvement over state-of-the-art models in terms of the optimality gap.
|
106,352 | 106,352 |
Pretraining Text Encoders with Adversarial Mixture of Training Signal
Generators
|
We present a new framework AMOS that pretrains text encoders with an
Adversarial learning curriculum via a Mixture Of Signals from multiple
auxiliary generators. Following ELECTRA-style pretraining, the main encoder is
trained as a discriminator to detect replaced tokens generated by auxiliary
masked language models (MLMs). Different from ELECTRA which trains one MLM as
the generator, we jointly train multiple MLMs of different sizes to provide
training signals at various levels of difficulty. To push the discriminator to
learn better with challenging replaced tokens, we learn mixture weights over
the auxiliary MLMs' outputs to maximize the discriminator loss by
backpropagating the gradient from the discriminator via Gumbel-Softmax. For
better pretraining efficiency, we propose a way to assemble multiple MLMs into
one unified auxiliary model. AMOS outperforms ELECTRA and recent
state-of-the-art pretrained models by about 1 point on the GLUE benchmark for
BERT base-sized models.
|
106,353 | 106,353 |
Composite Spatial Monte Carlo Integration Based on Generalized Least
Squares
|
Although evaluation of the expectations on the Ising model is essential in
various applications, this is frequently infeasible because of intractable
multiple summations (or integrations). Spatial Monte Carlo integration (SMCI)
is a sampling-based approximation, and can provide high-accuracy estimations
for such intractable expectations. To evaluate the expectation of a function of
variables in a specific region (called target region), SMCI considers a larger
region containing the target region (called sum region). In SMCI, the multiple
summation for the variables in the sum region is precisely executed, and that
in the outer region is evaluated by the sampling approximation such as the
standard Monte Carlo integration. It is guaranteed that the accuracy of the
SMCI estimator is monotonically improved as the size of the sum region
increases. However, a haphazard expansion of the sum region could cause a
combinatorial explosion. Therefore, we hope to improve the accuracy without
such region expansion. In this study, based on the theory of generalized least
squares, a new effective method is proposed by combining multiple SMCI
estimators. The validity of the proposed method is demonstrated theoretically
and numerically. The results indicate that the proposed method can be effective
in the inverse Ising problem (or Boltzmann machine learning).
|
106,354 | 106,354 |
mulEEG: A Multi-View Representation Learning on EEG Signals
|
Modeling effective representations using multiple views that positively
influence each other is challenging, and the existing methods perform poorly on
Electroencephalogram (EEG) signals for sleep-staging tasks. In this paper, we
propose a novel multi-view self-supervised method (mulEEG) for unsupervised EEG
representation learning. Our method attempts to effectively utilize the
complementary information available in multiple views to learn better
representations. We introduce diverse loss that further encourages
complementary information across multiple views. Our method with no access to
labels beats the supervised training while outperforming multi-view baseline
methods on transfer learning experiments carried out on sleep-staging tasks. We
posit that our method was able to learn better representations by using
complementary multi-views.
|
106,355 | 106,355 |
PALBERT: Teaching ALBERT to Ponder
|
Currently, pre-trained models can be considered the default choice for a wide
range of NLP tasks. Despite their SoTA results, there is practical evidence
that these models may require a different number of computing layers for
different input sequences, since evaluating all layers leads to overconfidence
on wrong predictions (namely overthinking). This problem can potentially be
solved by implementing adaptive computation time approaches, which were first
designed to improve inference speed. Recently proposed PonderNet may be a
promising solution for performing an early exit by treating the exit layers
index as a latent variable. However, the originally proposed exit criterion,
relying on sampling from trained posterior distribution on the probability of
exiting from i-th layer, introduces major variance in model outputs,
significantly reducing the resulting models performance. In this paper, we
propose Ponder ALBERT (PALBERT): an improvement to PonderNet with a novel
deterministic Q-exit criterion and a revisited model architecture. We compared
PALBERT with recent methods for performing an early exit. We observed that the
proposed changes can be considered significant improvements on the original
PonderNet architecture and outperform PABEE on a wide range of GLUE tasks. In
addition, we also performed an in-depth ablation study of the proposed
architecture to further understand Lambda layers and their performance.
|
106,356 | 106,356 |
Enhancing Semantic Code Search with Multimodal Contrastive Learning and
Soft Data Augmentation
|
Code search aims to retrieve the most semantically relevant code snippet for
a given natural language query. Recently, large-scale code pre-trained models
such as CodeBERT and GraphCodeBERT learn generic representations of source code
and have achieved substantial improvement on code search task. However, the
high-quality sequence-level representations of code snippets have not been
sufficiently explored. In this paper, we propose a new approach with multimodal
contrastive learning and soft data augmentation for code search. Multimodal
contrastive learning is used to pull together the representations of code-query
pairs and push apart the unpaired code snippets and queries. Moreover, data
augmentation is critical in contrastive learning for learning high-quality
representations. However, only semantic-preserving augmentations for source
code are considered in existing work. In this work, we propose to do soft data
augmentation by dynamically masking and replacing some tokens in code sequences
to generate code snippets that are similar but not necessarily
semantic-preserving as positive samples for paired queries. We conduct
extensive experiments to evaluate the effectiveness of our approach on a
large-scale dataset with six programming languages. The experimental results
show that our approach significantly outperforms the state-of-the-art methods.
We also adapt our techniques to several pre-trained models such as RoBERTa and
CodeBERT, and significantly boost their performance on the code search task.
|
106,357 | 106,357 |
Federated Learning from Only Unlabeled Data with
Class-Conditional-Sharing Clients
|
Supervised federated learning (FL) enables multiple clients to share the
trained model without sharing their labeled data. However, potential clients
might even be reluctant to label their own data, which could limit the
applicability of FL in practice. In this paper, we show the possibility of
unsupervised FL whose model is still a classifier for predicting class labels,
if the class-prior probabilities are shifted while the class-conditional
distributions are shared among the unlabeled data owned by the clients. We
propose federation of unsupervised learning (FedUL), where the unlabeled data
are transformed into surrogate labeled data for each of the clients, a modified
model is trained by supervised FL, and the wanted model is recovered from the
modified model. FedUL is a very general solution to unsupervised FL: it is
compatible with many supervised FL methods, and the recovery of the wanted
model can be theoretically guaranteed as if the data have been labeled.
Experiments on benchmark and real-world datasets demonstrate the effectiveness
of FedUL. Code is available at https://github.com/lunanbit/FedUL.
|
106,358 | 106,358 |
MBI-Net: A Non-Intrusive Multi-Branched Speech Intelligibility
Prediction Model for Hearing Aids
|
Improving the user's hearing ability to understand speech in noisy
environments is critical to the development of hearing aid (HA) devices. For
this, it is important to derive a metric that can fairly predict speech
intelligibility for HA users. A straightforward approach is to conduct a
subjective listening test and use the test results as an evaluation metric.
However, conducting large-scale listening tests is time-consuming and
expensive. Therefore, several evaluation metrics were derived as surrogates for
subjective listening test results. In this study, we propose a multi-branched
speech intelligibility prediction model (MBI-Net), for predicting the
subjective intelligibility scores of HA users. MBI-Net consists of two branches
of models, with each branch consisting of a hearing loss model, a cross-domain
feature extraction module, and a speech intelligibility prediction model, to
process speech signals from one channel. The outputs of the two branches are
fused through a linear layer to obtain predicted speech intelligibility scores.
Experimental results confirm the effectiveness of MBI-Net, which produces
higher prediction scores than the baseline system in Track 1 and Track 2 on the
Clarity Prediction Challenge 2022 dataset.
|
106,359 | 106,359 |
MTI-Net: A Multi-Target Speech Intelligibility Prediction Model
|
Recently, deep learning (DL)-based non-intrusive speech assessment models
have attracted great attention. Many studies report that these DL-based models
yield satisfactory assessment performance and good flexibility, but their
performance in unseen environments remains a challenge. Furthermore, compared
to quality scores, fewer studies elaborate deep learning models to estimate
intelligibility scores. This study proposes a multi-task speech intelligibility
prediction model, called MTI-Net, for simultaneously predicting human and
machine intelligibility measures. Specifically, given a speech utterance,
MTI-Net is designed to predict subjective listening test results and word error
rate (WER) scores. We also investigate several methods that can improve the
prediction performance of MTI-Net. First, we compare different features
(including low-level features and embeddings from self-supervised learning
(SSL) models) and prediction targets of MTI-Net. Second, we explore the effect
of transfer learning and multi-tasking learning on training MTI-Net. Finally,
we examine the potential advantages of fine-tuning SSL embeddings. Experimental
results demonstrate the effectiveness of using cross-domain features,
multi-task learning, and fine-tuning SSL embeddings. Furthermore, it is
confirmed that the intelligibility and WER scores predicted by MTI-Net are
highly correlated with the ground-truth scores.
|
106,360 | 106,360 |
Using Decision Tree as Local Interpretable Model in Autoencoder-based
LIME
|
Nowadays, deep neural networks are being used in many domains because of
their high accuracy results. However, they are considered as "black box", means
that they are not explainable for humans. On the other hand, in some tasks such
as medical, economic, and self-driving cars, users want the model to be
interpretable to decide if they can trust these results or not. In this work,
we present a modified version of an autoencoder-based approach for local
interpretability called ALIME. The ALIME itself is inspired by a famous method
called Local Interpretable Model-agnostic Explanations (LIME). LIME generates a
single instance level explanation by generating new data around the instance
and training a local linear interpretable model. ALIME uses an autoencoder to
weigh the new data around the sample. Nevertheless, the ALIME uses a linear
model as the interpretable model to be trained locally, just like the LIME.
This work proposes a new approach, which uses a decision tree instead of the
linear model, as the interpretable model. We evaluate the proposed model in
case of stability, local fidelity, and interpretability on different datasets.
Compared to ALIME, the experiments show significant results on stability and
local fidelity and improved results on interpretability.
|
106,361 | 106,361 |
Multi-Sample $\zeta$-mixup: Richer, More Realistic Synthetic Samples
from a $p$-Series Interpolant
|
Modern deep learning training procedures rely on model regularization
techniques such as data augmentation methods, which generate training samples
that increase the diversity of data and richness of label information. A
popular recent method, mixup, uses convex combinations of pairs of original
samples to generate new samples. However, as we show in our experiments, mixup
can produce undesirable synthetic samples, where the data is sampled off the
manifold and can contain incorrect labels. We propose $\zeta$-mixup, a
generalization of mixup with provably and demonstrably desirable properties
that allows convex combinations of $N \geq 2$ samples, leading to more
realistic and diverse outputs that incorporate information from $N$ original
samples by using a $p$-series interpolant. We show that, compared to mixup,
$\zeta$-mixup better preserves the intrinsic dimensionality of the original
datasets, which is a desirable property for training generalizable models.
Furthermore, we show that our implementation of $\zeta$-mixup is faster than
mixup, and extensive evaluation on controlled synthetic and 24 real-world
natural and medical image classification datasets shows that $\zeta$-mixup
outperforms mixup and traditional data augmentation techniques.
|
106,362 | 106,362 |
Enabling Deep Learning for All-in EDGE paradigm
|
Deep Learning-based models have been widely investigated, and they have
demonstrated significant performance on non-trivial tasks such as speech
recognition, image processing, and natural language understanding. However,
this is at the cost of substantial data requirements. Considering the
widespread proliferation of edge devices (e.g. Internet of Things devices) over
the last decade, Deep Learning in the edge paradigm, such as device-cloud
integrated platforms, is required to leverage its superior performance.
Moreover, it is suitable from the data requirements perspective in the edge
paradigm because the proliferation of edge devices has resulted in an explosion
in the volume of generated and collected data. However, there are difficulties
due to other requirements such as high computation, high latency, and high
bandwidth caused by Deep Learning applications in real-world scenarios. In this
regard, this survey paper investigates Deep Learning at the edge, its
architecture, enabling technologies, and model adaption techniques, where edge
servers and edge devices participate in deep learning training and inference.
For simplicity, we call this paradigm the All-in EDGE paradigm. Besides, this
paper presents the key performance metrics for Deep Learning at the All-in EDGE
paradigm to evaluate various deep learning techniques and choose a suitable
design. Moreover, various open challenges arising from the deployment of Deep
Learning at the All-in EDGE paradigm are identified and discussed.
|
106,363 | 106,363 |
Robust and Explainable Autoencoders for Unsupervised Time Series Outlier
Detection---Extended Version
|
Time series data occurs widely, and outlier detection is a fundamental
problem in data mining, which has numerous applications. Existing
autoencoder-based approaches deliver state-of-the-art performance on
challenging real-world data but are vulnerable to outliers and exhibit low
explainability. To address these two limitations, we propose robust and
explainable unsupervised autoencoder frameworks that decompose an input time
series into a clean time series and an outlier time series using autoencoders.
Improved explainability is achieved because clean time series are better
explained with easy-to-understand patterns such as trends and periodicities. We
provide insight into this by means of a post-hoc explainability analysis and
empirical studies. In addition, since outliers are separated from clean time
series iteratively, our approach offers improved robustness to outliers, which
in turn improves accuracy. We evaluate our approach on five real-world datasets
and report improvements over the state-of-the-art approaches in terms of
robustness and explainability.
This is an extended version of "Robust and Explainable Autoencoders for
Unsupervised Time Series Outlier Detection", to appear in IEEE ICDE 2022.
|
106,364 | 106,364 |
Domain Adaptation for Time-Series Classification to Mitigate Covariate
Shift
|
The performance of a machine learning model degrades when it is applied to
data from a similar but different domain than the data it has initially been
trained on. To mitigate this domain shift problem, domain adaptation (DA)
techniques search for an optimal transformation that converts the (current)
input data from a source domain to a target domain to learn a domain-invariant
representations that reduces domain discrepancy.
This paper proposes a novel supervised domain adaptation based on two steps.
First, we search for an optimal class-dependent transformation from the source
to the target domain from a few samples. We consider optimal transport methods
such as the earth mover distance with Laplacian regularization, Sinkhorn
transport and correlation alignment. Second, we use embedding similarity
techniques to select the corresponding transformation at inference. We use
correlation metrics and maximum mean discrepancy with higher-order moment
matching techniques. We conduct an extensive evaluation on time-series datasets
with domain shift including simulated and various online handwriting datasets
to demonstrate the performance.
|
106,365 | 106,365 |
Inference over radiative transfer models using variational and
expectation maximization methods
|
Earth observation from satellites offers the possibility to monitor our
planet with unprecedented accuracy. Radiative transfer models (RTMs) encode the
energy transfer through the atmosphere, and are used to model and understand
the Earth system, as well as to estimate the parameters that describe the
status of the Earth from satellite observations by inverse modeling. However,
performing inference over such simulators is a challenging problem. RTMs are
nonlinear, non-differentiable and computationally costly codes, which adds a
high level of difficulty in inference. In this paper, we introduce two
computational techniques to infer not only point estimates of biophysical
parameters but also their joint distribution. One of them is based on a
variational autoencoder approach and the second one is based on a Monte Carlo
Expectation Maximization (MCEM) scheme. We compare and discuss benefits and
drawbacks of each approach. We also provide numerical comparisons in synthetic
simulations and the real PROSAIL model, a popular RTM that combines land
vegetation leaf and canopy modeling. We analyze the performance of the two
approaches for modeling and inferring the distribution of three key biophysical
parameters for quantifying the terrestrial biosphere.
|
106,366 | 106,366 |
Offline Reinforcement Learning for Safer Blood Glucose Control in People
with Type 1 Diabetes
|
Hybrid closed loop systems represent the future of care for people with type
1 diabetes (T1D). These devices usually utilise simple control algorithms to
select the optimal insulin dose for maintaining blood glucose levels within a
healthy range. Online reinforcement learning (RL) has been utilised as a method
for further enhancing glucose control in these devices. Previous approaches
have been shown to reduce patient risk and improve time spent in the target
range when compared to classical control algorithms, but are prone to
instability in the learning process, often resulting in the selection of unsafe
actions. This work presents an evaluation of offline RL as a means for
developing clinically effective dosing policies without the need for patient
interaction. This paper examines the utility of BCQ, CQL and TD3-BC in managing
the blood glucose of nine virtual patients within the UVA/Padova glucose
dynamics simulator. When trained on less than a tenth of the data required by
online RL approaches, this work shows that offline RL can significantly
increase time in the healthy blood glucose range when compared to the strongest
state-of-art baseline. This is achieved without any associated increase in low
blood glucose events. Offline RL is also shown to be able to correct for common
and challenging scenarios such as incorrect bolus dosing, irregular meal
timings and sub-optimal training data.
|
106,367 | 106,367 |
Correcting Misproducted Speech using Spectrogram Inpainting
|
Learning a new language involves constantly comparing speech productions with
reference productions from the environment. Early in speech acquisition,
children make articulatory adjustments to match their caregivers' speech.
Grownup learners of a language tweak their speech to match the tutor reference.
This paper proposes a method to synthetically generate correct pronunciation
feedback given incorrect production. Furthermore, our aim is to generate the
corrected production while maintaining the speaker's original voice.
The system prompts the user to pronounce a phrase. The speech is recorded,
and the samples associated with the inaccurate phoneme are masked with zeros.
This waveform serves as an input to a speech generator, implemented as a deep
learning inpainting system with a U-net architecture, and trained to output a
reconstructed speech. The training set is composed of unimpaired proper speech
examples, and the generator is trained to reconstruct the original proper
speech. We evaluated the performance of our system on phoneme replacement of
minimal pair words of English as well as on children with pronunciation
disorders. Results suggest that human listeners slightly prefer our generated
speech over a smoothed replacement of the inaccurate phoneme with a production
of a different speaker.
|
106,368 | 106,368 |
Continual Inference: A Library for Efficient Online Inference with Deep
Neural Networks in PyTorch
|
We present Continual Inference, a Python library for implementing Continual
Inference Networks (CINs) in PyTorch, a class of Neural Networks designed
specifically for efficient inference in both online and batch processing
scenarios. We offer a comprehensive introduction and guide to CINs and their
implementation in practice, and provide best-practices and code examples for
composing complex modules for modern Deep Learning. Continual Inference is
readily downloadable via the Python Package Index and at
\url{www.github.com/lukashedegaard/continual-inference}.
|
106,369 | 106,369 |
Self supervised learning for robust voice cloning
|
Voice cloning is a difficult task which requires robust and informative
features incorporated in a high quality TTS system in order to effectively copy
an unseen speaker's voice. In our work, we utilize features learned in a
self-supervised framework via the Bootstrap Your Own Latent (BYOL) method,
which is shown to produce high quality speech representations when specific
audio augmentations are applied to the vanilla algorithm. We further extend the
augmentations in the training procedure to aid the resulting features to
capture the speaker identity and to make them robust to noise and acoustic
conditions. The learned features are used as pre-trained utterance-level
embeddings and as inputs to a Non-Attentive Tacotron based architecture, aiming
to achieve multispeaker speech synthesis without utilizing additional speaker
features. This method enables us to train our model in an unlabeled
multispeaker dataset as well as use unseen speaker embeddings to copy a
speaker's voice. Subjective and objective evaluations are used to validate the
proposed model, as well as the robustness to the acoustic conditions of the
target utterance.
|
106,370 | 106,370 |
Energy-Efficient Adaptive Machine Learning on IoT End-Nodes With
Class-Dependent Confidence
|
Energy-efficient machine learning models that can run directly on edge
devices are of great interest in IoT applications, as they can reduce network
pressure and response latency, and improve privacy. An effective way to obtain
energy-efficiency with small accuracy drops is to sequentially execute a set of
increasingly complex models, early-stopping the procedure for "easy" inputs
that can be confidently classified by the smallest models. As a stopping
criterion, current methods employ a single threshold on the output
probabilities produced by each model. In this work, we show that such a
criterion is sub-optimal for datasets that include classes of different
complexity, and we demonstrate a more general approach based on per-classes
thresholds. With experiments on a low-power end-node, we show that our method
can significantly reduce the energy consumption compared to the
single-threshold approach.
|
106,371 | 106,371 |
Machine Learning-Enabled IoT Security: Open Issues and Challenges Under
Advanced Persistent Threats
|
Despite its technological benefits, Internet of Things (IoT) has cyber
weaknesses due to the vulnerabilities in the wireless medium. Machine learning
(ML)-based methods are widely used against cyber threats in IoT networks with
promising performance. Advanced persistent threat (APT) is prominent for
cybercriminals to compromise networks, and it is crucial to long-term and
harmful characteristics. However, it is difficult to apply ML-based approaches
to identify APT attacks to obtain a promising detection performance due to an
extremely small percentage among normal traffic. There are limited surveys to
fully investigate APT attacks in IoT networks due to the lack of public
datasets with all types of APT attacks. It is worth to bridge the
state-of-the-art in network attack detection with APT attack detection in a
comprehensive review article. This survey article reviews the security
challenges in IoT networks and presents the well-known attacks, APT attacks,
and threat models in IoT systems. Meanwhile, signature-based, anomaly-based,
and hybrid intrusion detection systems are summarized for IoT networks. The
article highlights statistical insights regarding frequently applied ML-based
methods against network intrusion alongside the number of attacks types
detected. Finally, open issues and challenges for common network intrusion and
APT attacks are presented for future research.
|
106,372 | 106,372 |
Half-sibling regression meets exoplanet imaging: PSF modeling and
subtraction using a flexible, domain knowledge-driven, causal framework
|
High-contrast imaging of exoplanets hinges on powerful post-processing
methods to denoise the data and separate the signal of a companion from its
host star, which is typically orders of magnitude brighter. Existing
post-processing algorithms do not use all prior domain knowledge that is
available about the problem. We propose a new method that builds on our
understanding of the systematic noise and the causal structure of the
data-generating process. Our algorithm is based on a modified version of
half-sibling regression (HSR), a flexible denoising framework that combines
ideas from the fields of machine learning and causality. We adapt the method to
address the specific requirements of high-contrast exoplanet imaging data
obtained in pupil tracking mode. The key idea is to estimate the systematic
noise in a pixel by regressing the time series of this pixel onto a set of
causally independent, signal-free predictor pixels. We use regularized linear
models in this work; however, other (non-linear) models are also possible. In a
second step, we demonstrate how the HSR framework allows us to incorporate
observing conditions such as wind speed or air temperature as additional
predictors. When we apply our method to four data sets from the VLT/NACO
instrument, our algorithm provides a better false-positive fraction than
PCA-based PSF subtraction, a popular baseline method in the field.
Additionally, we find that the HSR-based method provides direct and accurate
estimates for the contrast of the exoplanets without the need to insert
artificial companions for calibration in the data sets. Finally, we present
first evidence that using the observing conditions as additional predictors can
improve the results. Our HSR-based method provides an alternative, flexible and
promising approach to the challenge of modeling and subtracting the stellar PSF
and systematic noise in exoplanet imaging data.
|
106,373 | 106,373 |
Few-Shot Forecasting of Time-Series with Heterogeneous Channels
|
Learning complex time series forecasting models usually requires a large
amount of data, as each model is trained from scratch for each task/data set.
Leveraging learning experience with similar datasets is a well-established
technique for classification problems called few-shot classification. However,
existing approaches cannot be applied to time-series forecasting because i)
multivariate time-series datasets have different channels and ii) forecasting
is principally different from classification. In this paper we formalize the
problem of few-shot forecasting of time-series with heterogeneous channels for
the first time. Extending recent work on heterogeneous attributes in vector
data, we develop a model composed of permutation-invariant deep set-blocks
which incorporate a temporal embedding. We assemble the first meta-dataset of
40 multivariate time-series datasets and show through experiments that our
model provides a good generalization, outperforming baselines carried over from
simpler scenarios that either fail to learn across tasks or miss temporal
information.
|
106,374 | 106,374 |
Video Diffusion Models
|
Generating temporally coherent high fidelity video is an important milestone
in generative modeling research. We make progress towards this milestone by
proposing a diffusion model for video generation that shows very promising
initial results. Our model is a natural extension of the standard image
diffusion architecture, and it enables jointly training from image and video
data, which we find to reduce the variance of minibatch gradients and speed up
optimization. To generate long and higher resolution videos we introduce a new
conditional sampling technique for spatial and temporal video extension that
performs better than previously proposed methods. We present the first results
on a large text-conditioned video generation task, as well as state-of-the-art
results on an established unconditional video generation benchmark.
Supplementary material is available at https://video-diffusion.github.io/
|
106,375 | 106,375 |
BERTuit: Understanding Spanish language in Twitter through a native
transformer
|
The appearance of complex attention-based language models such as BERT,
Roberta or GPT-3 has allowed to address highly complex tasks in a plethora of
scenarios. However, when applied to specific domains, these models encounter
considerable difficulties. This is the case of Social Networks such as Twitter,
an ever-changing stream of information written with informal and complex
language, where each message requires careful evaluation to be understood even
by humans given the important role that context plays. Addressing tasks in this
domain through Natural Language Processing involves severe challenges. When
powerful state-of-the-art multilingual language models are applied to this
scenario, language specific nuances use to get lost in translation. To face
these challenges we present \textbf{BERTuit}, the larger transformer proposed
so far for Spanish language, pre-trained on a massive dataset of 230M Spanish
tweets using RoBERTa optimization. Our motivation is to provide a powerful
resource to better understand Spanish Twitter and to be used on applications
focused on this social network, with special emphasis on solutions devoted to
tackle the spreading of misinformation in this platform. BERTuit is evaluated
on several tasks and compared against M-BERT, XLM-RoBERTa and XLM-T, very
competitive multilingual transformers. The utility of our approach is shown
with applications, in this case: a zero-shot methodology to visualize groups of
hoaxes and profiling authors spreading disinformation.
Misinformation spreads wildly on platforms such as Twitter in languages other
than English, meaning performance of transformers may suffer when transferred
outside English speaking communities.
|
106,376 | 106,376 |
Jacobian Norm for Unsupervised Source-Free Domain Adaptation
|
Unsupervised Source (data) Free domain adaptation (USFDA) aims to transfer
knowledge from a well-trained source model to a related but unlabeled target
domain. In such a scenario, all conventional adaptation methods that require
source data fail. To combat this challenge, existing USFDAs turn to transfer
knowledge by aligning the target feature to the latent distribution hidden in
the source model. However, such information is naturally limited. Thus, the
alignment in such a scenario is not only difficult but also insufficient, which
degrades the target generalization performance. To relieve this dilemma in
current USFDAs, we are motivated to explore a new perspective to boost their
performance. For this purpose and gaining necessary insight, we look back upon
the origin of the domain adaptation and first theoretically derive a new-brand
target generalization error bound based on the model smoothness. Then,
following the theoretical insight, a general and model-smoothness-guided
Jacobian norm (JN) regularizer is designed and imposed on the target domain to
mitigate this dilemma. Extensive experiments are conducted to validate its
effectiveness. In its implementation, just with a few lines of codes added to
the existing USFDAs, we achieve superior results on various benchmark datasets.
|
106,377 | 106,377 |
DynLight: Realize dynamic phase duration with multi-level traffic signal
control
|
Adopting reinforcement learning (RL) for traffic signal control (TSC) is
increasingly popular, and RL has become a promising solution for traffic signal
control. However, several challenges still need to be overcome. Firstly, most
RL methods use fixed action duration and select the green phase for the next
state, which makes the phase duration less dynamic and flexible. Secondly, the
phase sequence of RL methods can be arbitrary, affecting the real-world
deployment which may require a cyclical phase structure. Lastly, the average
travel time and throughput are not fair metrics to evaluate TSC performance. To
address these challenges, we propose a multi-level traffic signal control
framework, DynLight, which uses an optimization method Max-QueueLength (M-QL)
to determine the phase and uses a deep Q-network to determine the duration of
the corresponding phase. Based on DynLight, we further propose DynLight-C which
adopts a well-trained deep Q-network of DynLight and replace M-QL with a
cyclical control policy that actuates a set of phases in fixed cyclical order
to realize cyclical phase structure. Comprehensive experiments on multiple
real-world datasets demonstrate that DynLight achieves a new state-of-the-art.
Furthermore, the deep Q-network of DynLight can learn well on determining the
phase duration and DynLight-C demonstrates high performance for deployment.
|
106,378 | 106,378 |
Solving ImageNet: a Unified Scheme for Training any Backbone to Top
Results
|
ImageNet serves as the primary dataset for evaluating the quality of
computer-vision models. The common practice today is training each architecture
with a tailor-made scheme, designed and tuned by an expert. In this paper, we
present a unified scheme for training any backbone on ImageNet. The scheme,
named USI (Unified Scheme for ImageNet), is based on knowledge distillation and
modern tricks. It requires no adjustments or hyper-parameters tuning between
different models, and is efficient in terms of training times. We test USI on a
wide variety of architectures, including CNNs, Transformers, Mobile-oriented
and MLP-only. On all models tested, USI outperforms previous state-of-the-art
results. Hence, we are able to transform training on ImageNet from an
expert-oriented task to an automatic seamless routine. Since USI accepts any
backbone and trains it to top results, it also enables to perform methodical
comparisons, and identify the most efficient backbones along the speed-accuracy
Pareto curve. Implementation is available
at:https://github.com/Alibaba-MIIL/Solving_ImageNet
|
106,379 | 106,379 |
Delta Keyword Transformer: Bringing Transformers to the Edge through
Dynamically Pruned Multi-Head Self-Attention
|
Multi-head self-attention forms the core of Transformer networks. However,
their quadratically growing complexity with respect to the input sequence
length impedes their deployment on resource-constrained edge devices. We
address this challenge by proposing a dynamic pruning method, which exploits
the temporal stability of data across tokens to reduce inference cost. The
threshold-based method only retains significant differences between the
subsequent tokens, effectively reducing the number of multiply-accumulates, as
well as the internal tensor data sizes. The approach is evaluated on the Google
Speech Commands Dataset for keyword spotting, and the performance is compared
against the baseline Keyword Transformer. Our experiments show that we can
reduce ~80% of operations while maintaining the original 98.4% accuracy.
Moreover, a reduction of ~87-94% operations can be achieved when only degrading
the accuracy by 1-4%, speeding up the multi-head self-attention inference by a
factor of ~7.5-16.
|
106,380 | 106,380 |
Optimizing the Long-Term Behaviour of Deep Reinforcement Learning for
Pushing and Grasping
|
We investigate the "Visual Pushing for Grasping" (VPG) system by Zeng et al.
and the "Hourglass" system by Ewerton et al., an evolution of the former. The
focus of our work is the investigation of the capabilities of both systems to
learn long-term rewards and policies. Zeng et al. original task only needs a
limited amount of foresight. Ewerton et al. attain their best performance using
an agent which only takes the most immediate action under consideration. We are
interested in the ability of their models and training algorithms to accurately
predict long-term Q-Values. To evaluate this ability, we design a new bin
sorting task and reward function. Our task requires agents to accurately
estimate future rewards and therefore use high discount factors in their
Q-Value calculation. We investigate the behaviour of an adaptation of the VPG
training algorithm on our task. We show that this adaptation can not accurately
predict the required long-term action sequences. In addition to the limitations
identified by Ewerton et al., it suffers from the known Deep Q-Learning problem
of overestimated Q-Values. In an effort to solve our task, we turn to the
Hourglass models and combine them with the Double Q-Learning approach. We show
that this approach enables the models to accurately predict long-term action
sequences when trained with large discount factors. Our results show that the
Double Q-Learning technique is essential for training with very high discount
factors, as the models Q-Value predictions diverge otherwise. We also
experiment with different approaches for discount factor scheduling, loss
calculation and exploration procedures. Our results show that the latter
factors do not visibly influence the model's performance for our task.
|
106,381 | 106,381 |
Position-based Prompting for Health Outcome Generation
|
Probing Pre-trained Language Models (PLMs) using prompts has indirectly
implied that language models (LMs) can be treated as knowledge bases. To this
end, this phenomena has been effective especially when these LMs are fine-tuned
towards not just data of a specific domain, but also to the style or linguistic
pattern of the prompts themselves. We observe that, satisfying a particular
linguistic pattern in prompts is an unsustainable constraint that unnecessarily
lengthens the probing task, especially because, they are often manually
designed and the range of possible prompt template patterns can vary depending
on the prompting objective and domain. We therefore explore an idea of using a
position-attention mechanism to capture positional information of each word in
a prompt relative to the mask to be filled, hence avoiding the need to
re-construct prompts when the prompts linguistic pattern changes. Using our
approach, we demonstrate the ability of eliciting answers to rare prompt
templates (in a case study on health outcome generation) such as Postfix and
Mixed patterns whose missing information is respectively at the start and in
multiple random places of the prompt. More so, using various biomedical PLMs,
our approach consistently outperforms a baseline in which the default mask
language model (MLM) representation is used to predict masked tokens.
|
106,382 | 106,382 |
Covariance matrix preparation for quantum principal component analysis
|
Principal component analysis (PCA) is a dimensionality reduction method in
data analysis that involves diagonalizing the covariance matrix of the dataset.
Recently, quantum algorithms have been formulated for PCA based on
diagonalizing a density matrix. These algorithms assume that the covariance
matrix can be encoded in a density matrix, but a concrete protocol for this
encoding has been lacking. Our work aims to address this gap. Assuming
amplitude encoding of the data, with the data given by the ensemble $\{p_i,|
\psi_i \rangle\}$, then one can easily prepare the ensemble average density
matrix $\overline{\rho} = \sum_i p_i |\psi_i\rangle \langle \psi_i |$. We first
show that $\overline{\rho}$ is precisely the covariance matrix whenever the
dataset is centered. For quantum datasets, we exploit global phase symmetry to
argue that there always exists a centered dataset consistent with
$\overline{\rho}$, and hence $\overline{\rho}$ can always be interpreted as a
covariance matrix. This provides a simple means for preparing the covariance
matrix for arbitrary quantum datasets or centered classical datasets. For
uncentered classical datasets, our method is so-called "PCA without centering",
which we interpret as PCA on a symmetrized dataset. We argue that this closely
corresponds to standard PCA, and we derive equations and inequalities that
bound the deviation of the spectrum obtained with our method from that of
standard PCA. We numerically illustrate our method for the MNIST handwritten
digit dataset. We also argue that PCA on quantum datasets is natural and
meaningful, and we numerically implement our method for molecular ground-state
datasets.
|
106,383 | 106,383 |
Generalised Latent Assimilation in Heterogeneous Reduced Spaces with
Machine Learning Surrogate Models
|
Reduced-order modelling and low-dimensional surrogate models generated using
machine learning algorithms have been widely applied in high-dimensional
dynamical systems to improve the algorithmic efficiency. In this paper, we
develop a system which combines reduced-order surrogate models with a novel
data assimilation (DA) technique used to incorporate real-time observations
from different physical spaces. We make use of local smooth surrogate functions
which link the space of encoded system variables and the one of current
observations to perform variational DA with a low computational cost. The new
system, named Generalised Latent Assimilation can benefit both the efficiency
provided by the reduced-order modelling and the accuracy of data assimilation.
A theoretical analysis of the difference between surrogate and original
assimilation cost function is also provided in this paper where an upper bound,
depending on the size of the local training set, is given. The new approach is
tested on a high-dimensional CFD application of a two-phase liquid flow with
non-linear observation operators that current Latent Assimilation methods can
not handle. Numerical results demonstrate that the proposed assimilation
approach can significantly improve the reconstruction and prediction accuracy
of the deep learning surrogate model which is nearly 1000 times faster than the
CFD simulation.
|
106,384 | 106,384 |
On the Effectiveness of Pretrained Models for API Learning
|
Developers frequently use APIs to implement certain functionalities, such as
parsing Excel Files, reading and writing text files line by line, etc.
Developers can greatly benefit from automatic API usage sequence generation
based on natural language queries for building applications in a faster and
cleaner manner. Existing approaches utilize information retrieval models to
search for matching API sequences given a query or use RNN-based
encoder-decoder to generate API sequences. As it stands, the first approach
treats queries and API names as bags of words. It lacks deep comprehension of
the semantics of the queries. The latter approach adapts a neural language
model to encode a user query into a fixed-length context vector and generate
API sequences from the context vector.
We want to understand the effectiveness of recent Pre-trained Transformer
based Models (PTMs) for the API learning task. These PTMs are trained on large
natural language corpora in an unsupervised manner to retain contextual
knowledge about the language and have found success in solving similar Natural
Language Processing (NLP) problems. However, the applicability of PTMs has not
yet been explored for the API sequence generation task. We use a dataset that
contains 7 million annotations collected from GitHub to evaluate the PTMs
empirically. This dataset was also used to assess previous approaches. Based on
our results, PTMs generate more accurate API sequences and outperform other
related methods by around 11%. We have also identified two different
tokenization approaches that can contribute to a significant boost in PTMs'
performance for the API sequence generation task.
|
106,385 | 106,385 |
Multi-Task Distributed Learning using Vision Transformer with Random
Patch Permutation
|
The widespread application of artificial intelligence in health research is
currently hampered by limitations in data availability. Distributed learning
methods such as federated learning (FL) and shared learning (SL) are introduced
to solve this problem as well as data management and ownership issues with
their different strengths and weaknesses. The recent proposal of federated
split task-agnostic (FeSTA) learning tries to reconcile the distinct merits of
FL and SL by enabling the multi-task collaboration between participants through
Vision Transformer (ViT) architecture, but they suffer from higher
communication overhead. To address this, here we present a multi-task
distributed learning using ViT with random patch permutation. Instead of using
a CNN based head as in FeSTA, p-FeSTA adopts a randomly permuting simple patch
embedder, improving the multi-task learning performance without sacrificing
privacy. Experimental results confirm that the proposed method significantly
enhances the benefit of multi-task collaboration, communication efficiency, and
privacy preservation, shedding light on practical multi-task distributed
learning in the field of medical imaging.
|
106,386 | 106,386 |
Survey on Automated Short Answer Grading with Deep Learning: from Word
Embeddings to Transformers
|
Automated short answer grading (ASAG) has gained attention in education as a
means to scale educational tasks to the growing number of students. Recent
progress in Natural Language Processing and Machine Learning has largely
influenced the field of ASAG, of which we survey the recent research
advancements. We complement previous surveys by providing a comprehensive
analysis of recently published methods that deploy deep learning approaches. In
particular, we focus our analysis on the transition from hand engineered
features to representation learning approaches, which learn representative
features for the task at hand automatically from large corpora of data. We
structure our analysis of deep learning methods along three categories: word
embeddings, sequential models, and attention-based methods. Deep learning
impacted ASAG differently than other fields of NLP, as we noticed that the
learned representations alone do not contribute to achieve the best results,
but they rather show to work in a complementary way with hand-engineered
features. The best performance are indeed achieved by methods that combine the
carefully hand-engineered features with the power of the semantic descriptions
provided by the latest models, like transformers architectures. We identify
challenges and provide an outlook on research direction that can be addressed
in the future
|
106,387 | 106,387 |
AI-aided Traffic Control Scheme for M2M Communications in the Internet
of Vehicles
|
Due to the rapid growth of data transmissions in internet of vehicles (IoV),
finding schemes that can effectively alleviate access congestion has become an
important issue. Recently, many traffic control schemes have been studied.
Nevertheless, the dynamics of traffic and the heterogeneous requirements of
different IoV applications are not considered in most existing studies, which
is significant for the random access resource allocation. In this paper, we
consider a hybrid traffic control scheme and use proximal policy optimization
(PPO) method to tackle it. Firstly, IoV devices are divided into various
classes based on delay characteristics. The target of maximizing the successful
transmission of packets with the success rate constraint is established. Then,
the optimization objective is transformed into a markov decision process (MDP)
model. Finally, the access class barring (ACB) factors are obtained based on
the PPO method to maximize the number of successful access devices. The
performance of the proposal algorithm in respect of successful events and delay
compared to existing schemes is verified by simulations.
|
106,388 | 106,388 |
Interval Bound Propagation$\unicode{x2013}$aided
Few$\unicode{x002d}$shot Learning
|
Few-shot learning aims to transfer the knowledge acquired from training on a
diverse set of tasks, from a given task distribution, to generalize to unseen
tasks, from the same distribution, with a limited amount of labeled data. The
underlying requirement for effective few-shot generalization is to learn a good
representation of the task manifold. One way to encourage this is to preserve
local neighborhoods in the feature space learned by the few-shot learner. To
this end, we introduce the notion of interval bounds from the provably robust
training literature to few-shot learning. The interval bounds are used to
characterize neighborhoods around the training tasks. These neighborhoods can
then be preserved by minimizing the distance between a task and its respective
bounds. We further introduce a novel strategy to artificially form new tasks
for training by interpolating between the available tasks and their respective
interval bounds, to aid in cases with a scarcity of tasks. We apply our
framework to both model-agnostic meta-learning as well as prototype-based
metric-learning paradigms. The efficacy of our proposed approach is evident
from the improved performance on several datasets from diverse domains in
comparison to a sizable number of recent competitors.
|
106,389 | 106,389 |
Distributed Reinforcement Learning for Robot Teams: A Review
|
Purpose of review: Recent advances in sensing, actuation, and computation
have opened the door to multi-robot systems consisting of hundreds/thousands of
robots, with promising applications to automated manufacturing, disaster
relief, harvesting, last-mile delivery, port/airport operations, or search and
rescue. The community has leveraged model-free multi-agent reinforcement
learning (MARL) to devise efficient, scalable controllers for multi-robot
systems (MRS). This review aims to provide an analysis of the state-of-the-art
in distributed MARL for multi-robot cooperation.
Recent findings: Decentralized MRS face fundamental challenges, such as
non-stationarity and partial observability. Building upon the "centralized
training, decentralized execution" paradigm, recent MARL approaches include
independent learning, centralized critic, value decomposition, and
communication learning approaches. Cooperative behaviors are demonstrated
through AI benchmarks and fundamental real-world robotic capabilities such as
multi-robot motion/path planning.
Summary: This survey reports the challenges surrounding decentralized
model-free MARL for multi-robot cooperation and existing classes of approaches.
We present benchmarks and robotic applications along with a discussion on
current open avenues for research.
|
106,390 | 106,390 |
Temporal Alignment for History Representation in Reinforcement Learning
|
Environments in Reinforcement Learning are usually only partially observable.
To address this problem, a possible solution is to provide the agent with
information about the past. However, providing complete observations of
numerous steps can be excessive. Inspired by human memory, we propose to
represent history with only important changes in the environment and, in our
approach, to obtain automatically this representation using self-supervision.
Our method (TempAl) aligns temporally-close frames, revealing a general, slowly
varying state of the environment. This procedure is based on contrastive loss,
which pulls embeddings of nearby observations to each other while pushing away
other samples from the batch. It can be interpreted as a metric that captures
the temporal relations of observations. We propose to combine both common
instantaneous and our history representation and we evaluate TempAl on all
available Atari games from the Arcade Learning Environment. TempAl surpasses
the instantaneous-only baseline in 35 environments out of 49. The source code
of the method and of all the experiments is available at
https://github.com/htdt/tempal.
|
106,391 | 106,391 |
Visualizing Deep Neural Networks with Topographic Activation Maps
|
Machine Learning with Deep Neural Networks (DNNs) has become a successful
tool in solving tasks across various fields of application. The success of DNNs
is strongly connected to their high complexity in terms of the number of
network layers or of neurons in each layer, which severely complicates to
understand how DNNs solve their learned task. To improve the explainability of
DNNs, we adapt methods from neuroscience because this field has a rich
experience in analyzing complex and opaque systems. In this work, we draw
inspiration from how neuroscience uses topographic maps to visualize the
activity of the brain when it performs certain tasks. Transferring this
approach to DNNs can help to visualize and understand their internal processes
more intuitively, too. However, the inner structures of brains and DNNs differ
substantially. Therefore, to be able to visualize activations of neurons in
DNNs as topographic maps, we research techniques to layout the neurons in a
two-dimensional space in which neurons of similar activity are in the vicinity
of each other. In this work, we introduce and compare different methods to
obtain a topographic layout of the neurons in a network layer. Moreover, we
demonstrate how to use the resulting topographic activation maps to identify
errors or encoded biases in DNNs or data sets. Our novel visualization
technique improves the transparency of DNN-based algorithmic decision-making
systems and is accessible to a broad audience because topographic maps are
intuitive to interpret without expert-knowledge in Machine Learning.
|
106,392 | 106,392 |
FedADMM: A Robust Federated Deep Learning Framework with Adaptivity to
System Heterogeneity
|
Federated Learning (FL) is an emerging framework for distributed processing
of large data volumes by edge devices subject to limited communication
bandwidths, heterogeneity in data distributions and computational resources, as
well as privacy considerations. In this paper, we introduce a new FL protocol
termed FedADMM based on primal-dual optimization. The proposed method leverages
dual variables to tackle statistical heterogeneity, and accommodates system
heterogeneity by tolerating variable amount of work performed by clients.
FedADMM maintains identical communication costs per round as FedAvg/Prox, and
generalizes them via the augmented Lagrangian. A convergence proof is
established for nonconvex objectives, under no restrictions in terms of data
dissimilarity or number of participants per round of the algorithm. We
demonstrate the merits through extensive experiments on real datasets, under
both IID and non-IID data distributions across clients. FedADMM consistently
outperforms all baseline methods in terms of communication efficiency, with the
number of rounds needed to reach a prescribed accuracy reduced by up to 87%.
The algorithm effectively adapts to heterogeneous data distributions through
the use of dual variables, without the need for hyperparameter tuning, and its
advantages are more pronounced in large-scale systems.
|
106,393 | 106,393 |
RF Signal Transformation and Classification using Deep Neural Networks
|
Deep neural networks (DNNs) designed for computer vision and natural language
processing tasks cannot be directly applied to the radio frequency (RF)
datasets. To address this challenge, we propose to convert the raw RF data to
data types that are suitable for off-the-shelf DNNs by introducing a
convolutional transform technique. In addition, we propose a simple 5-layer
convolutional neural network architecture (CONV-5) that can operate with raw RF
I/Q data without any transformation. Further, we put forward an RF dataset,
referred to as RF1024, to facilitate future RF research. RF1024 consists of 8
different RF modulation classes with each class having 1000/200 training/test
samples. Each sample of the RF1024 dataset contains 1024 complex I/Q values.
Lastly, the experiments are performed on the RadioML2016 and RF1024 datasets to
demonstrate the improved classification performance.
|
106,394 | 106,394 |
Adaptive Spike-Like Representation of EEG Signals for Sleep Stages
Scoring
|
Recently there has seen promising results on automatic stage scoring by
extracting spatio-temporal features from electroencephalogram (EEG). Such
methods entail laborious manual feature engineering and domain knowledge. In
this study, we propose an adaptive scheme to probabilistically encode, filter
and accumulate the input signals and weight the resultant features by the
half-Gaussian probabilities of signal intensities. The adaptive representations
are subsequently fed into a transformer model to automatically mine the
relevance between features and corresponding stages. Extensive experiments on
the largest public dataset against state-of-the-art methods validate the
effectiveness of our proposed method and reveal promising future directions.
|
106,395 | 106,395 |
Faster algorithms for learning to link, align sequences, and price
two-part tariffs
|
Data-driven algorithm configuration is a promising, learning-based approach
for beyond worst-case analysis of algorithms with tunable parameters. An
important open problem is the design of efficient data-driven algorithms for
algorithm families with more than one parameter. In this work we provide
algorithms for efficient (output-polynomial) multidimensional parameter tuning,
i.e. for families with a small constant number of parameters, for three very
different combinatorial problems -- linkage-based clustering, dynamic
programming for sequence alignment, and auction design for two-part tariff
schemes. We extend the single-parameter clustering algorithm of Balcan et al.
2020 arXiv:1907.00533 to multiple parameters and to the sequence alignment
problem by proposing an execution graph which compactly represents all the
states the algorithm could attain for all possible parameter values. A key
problem-specific challenge is to efficiently compute how the partition of the
parameter space (into regions with unique algorithmic states) changes with a
single algorithmic step. We give algorithms which improve on the runtime of
previously best known results for linkage-based clustering, sequence alignment
and two-part tariff pricing.
|
106,396 | 106,396 |
Improving Urban Mobility: using artificial intelligence and new
technologies to connect supply and demand
|
As the demand for mobility in our society seems to increase, the various
issues centered on urban mobility are among those that worry most city
inhabitants in this planet. For instance, how to go from A to B in an efficient
(but also less stressful) way? These questions and concerns have not changed
even during the covid-19 pandemic; on the contrary, as the current stand,
people who are avoiding public transportation are only contributing to an
increase in the vehicular traffic. The are of intelligent transportation
systems (ITS) aims at investigating how to employ information and communication
technologies to problems related to transportation. This may mean monitoring
and managing the infrastructure (e.g., traffic roads, traffic signals, etc.).
However, currently, ITS is also targeting the management of demand. In this
panorama, artificial intelligence plays an important role, especially with the
advances in machine learning that translates in the use of computational
vision, connected and autonomous vehicles, agent-based simulation, among
others. In the present work, a survey of several works developed by our group
are discussed in a holistic perspective, i.e., they cover not only the supply
side (as commonly found in ITS works), but also the demand side, and, in an
novel perspective, the integration of both.
|
106,397 | 106,397 |
A Pathology-Based Machine Learning Method to Assist in Epithelial
Dysplasia Diagnosis
|
The Epithelial Dysplasia (ED) is a tissue alteration commonly present in
lesions preceding oral cancer, being its presence one of the most important
factors in the progression toward carcinoma. This study proposes a method to
design a low computational cost classification system to support the detection
of dysplastic epithelia, contributing to reduce the variability of pathologist
assessments. We employ a multilayer artificial neural network (MLP-ANN) and
defining the regions of the epithelium to be assessed based on the knowledge of
the pathologist. The performance of the proposed solution was statistically
evaluated. The implemented MLP-ANN presented an average accuracy of 87%, with a
variability much inferior to that obtained from three trained evaluators.
Moreover, the proposed solution led to results which are very close to those
obtained using a convolutional neural network (CNN) implemented by transfer
learning, with 100 times less computational complexity. In conclusion, our
results show that a simple neural network structure can lead to a performance
equivalent to that of much more complex structures, which are routinely used in
the literature.
|
106,398 | 106,398 |
An optimized hybrid solution for IoT based lifestyle disease
classification using stress data
|
Stress, anxiety, and nervousness are all high-risk health states in everyday
life. Previously, stress levels were determined by speaking with people and
gaining insight into what they had experienced recently or in the past.
Typically, stress is caused by an incidence that occurred a long time ago, but
sometimes it is triggered by unknown factors. This is a challenging and complex
task, but recent research advances have provided numerous opportunities to
automate it. The fundamental features of most of these techniques are electro
dermal activity (EDA) and heart rate values (HRV). We utilized an accelerometer
to measure body motions to solve this challenge. The proposed novel method
employs a test that measures a subject's electrocardiogram (ECG), galvanic skin
values (GSV), HRV values, and body movements in order to provide a low-cost and
time-saving solution for detecting stress lifestyle disease in modern times
using cyber physical systems. This study provides a new hybrid model for
lifestyle disease classification that decreases execution time while picking
the best collection of characteristics and increases classification accuracy.
The developed approach is capable of dealing with the class imbalance problem
by using WESAD (wearable stress and affect dataset) dataset. The new model uses
the Grid search (GS) method to select an optimized set of hyper parameters, and
it uses a combination of the Correlation coefficient based Recursive feature
elimination (CoC-RFE) method for optimal feature selection and gradient
boosting as an estimator to classify the dataset, which achieves high accuracy
and helps to provide smart, accurate, and high-quality healthcare systems. To
demonstrate the validity and utility of the proposed methodology, its
performance is compared to those of other well-established machine learning
models.
|
106,399 | 106,399 |
Learning to Compose Soft Prompts for Compositional Zero-Shot Learning
|
We introduce compositional soft prompting (CSP), a parameter-efficient
learning technique to improve the zero-shot compositionality of large-scale
pretrained vision-language models (VLMs) without the overhead of fine-tuning
the entire model. VLMs can represent arbitrary classes as natural language
prompts in their flexible text encoders but they underperform state-of-the-art
methods on compositional zero-shot benchmark tasks. To improve VLMs, we propose
a novel form of soft prompting. We treat the attributes and objects that are
composed to define classes as learnable tokens of vocabulary and tune them on
multiple prompt compositions. During inference, we recompose the learned
attribute-object vocabulary in new combinations and show that CSP outperforms
the original VLM on benchmark datasets by an average of 14.7 percentage points
of accuracy. CSP also achieves new state-of-the-art accuracies on two out of
three benchmark datasets, while only fine-tuning a small number of parameters.
Further, we show that CSP improves generalization to higher-order
attribute-attribute-object compositions and combinations of pretrained
attributes and fine-tuned objects.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.