Unnamed: 0.1
int64 0
113k
| Unnamed: 0
float64 0
113k
⌀ | title
stringlengths 7
246
| abstract
stringlengths 6
3.31k
|
|---|---|---|---|
110,800 | 110,800 | Weakly Supervised Representation Learning with Sparse Perturbations | The theory of representation learning aims to build methods that provably
invert the data generating process with minimal domain knowledge or any source
of supervision. Most prior approaches require strong distributional assumptions
on the latent variables and weak supervision (auxiliary information such as
timestamps) to provide provable identification guarantees. In this work, we
show that if one has weak supervision from observations generated by sparse
perturbations of the latent variables--e.g. images in a reinforcement learning
environment where actions move individual sprites--identification is achievable
under unknown continuous latent distributions. We show that if the
perturbations are applied only on mutually exclusive blocks of latents, we
identify the latents up to those blocks. We also show that if these
perturbation blocks overlap, we identify latents up to the smallest blocks
shared across perturbations. Consequently, if there are blocks that intersect
in one latent variable only, then such latents are identified up to permutation
and scaling. We propose a natural estimation procedure based on this theory and
illustrate it on low-dimensional synthetic and image-based experiments.
|
110,801 | 110,801 | Robustness to Label Noise Depends on the Shape of the Noise Distribution
in Feature Space | Machine learning classifiers have been demonstrated, both empirically and
theoretically, to be robust to label noise under certain conditions -- notably
the typical assumption is that label noise is independent of the features given
the class label. We provide a theoretical framework that generalizes beyond
this typical assumption by modeling label noise as a distribution over feature
space. We show that both the scale and the shape of the noise distribution
influence the posterior likelihood; and the shape of the noise distribution has
a stronger impact on classification performance if the noise is concentrated in
feature space where the decision boundary can be moved. For the special case of
uniform label noise (independent of features and the class label), we show that
the Bayes optimal classifier for $c$ classes is robust to label noise until the
ratio of noisy samples goes above $\frac{c-1}{c}$ (e.g. 90% for 10 classes),
which we call the tipping point. However, for the special case of
class-dependent label noise (independent of features given the class label),
the tipping point can be as low as 50%. Most importantly, we show that when the
noise distribution targets decision boundaries (label noise is directly
dependent on feature space), classification robustness can drop off even at a
small scale of noise. Even when evaluating recent label-noise mitigation
methods we see reduced accuracy when label noise is dependent on features.
These findings explain why machine learning often handles label noise well if
the noise distribution is uniform in feature-space; yet it also points to the
difficulty of overcoming label noise when it is concentrated in a region of
feature space where a decision boundary can move.
|
110,802 | 110,802 | Comparing Conventional and Deep Feature Models for Classifying Fundus
Photography of Hemorrhages | Diabetic retinopathy is an eye-related pathology creating abnormalities and
causing visual impairment, proper treatment of which requires identifying
irregularities. This research uses a hemorrhage detection method and compares
classification of conventional and deep features. Especially, method identifies
hemorrhage connected with blood vessels or reside at retinal border and
reported challenging. Initially, adaptive brightness adjustment and contrast
enhancement rectify degraded images. Prospective locations of hemorrhages are
estimated by a Gaussian matched filter, entropy thresholding, and morphological
operation. Hemorrhages are segmented by a novel technique based on regional
variance of intensities. Features are then extracted by conventional methods
and deep models for training support vector machines, and results evaluated.
Evaluation metrics for each model are promising, but findings suggest that
comparatively, deep models are more effective than conventional features.
|
110,803 | 110,803 | Super-resolving 2D stress tensor field conserving equilibrium
constraints using physics informed U-Net | In a finite element analysis, using a large number of grids is important to
obtain accurate results, but is a resource-consuming task. Aiming to real-time
simulation and optimization, it is desired to obtain fine grid analysis results
within a limited resource. This paper proposes a super-resolution method that
predicts a stress tensor field in a high-resolution from low-resolution contour
plots by utilizing a U-Net-based neural network which is called PI-UNet. In
addition, the proposed model minimizes the residual of the equilibrium
constraints so that it outputs a physically reasonable solution. The proposed
network is trained with FEM results of simple shapes, and is validated with a
complicated realistic shape to evaluate generalization capability. Although
ESRGAN is a standard model for image super-resolution, the proposed U-Net based
model outperforms ESRGAN model in the stress tensor prediction task.
|
110,804 | 110,804 | Predictive Multiplicity in Probabilistic Classification | For a prediction task, there may exist multiple models that perform almost
equally well. This multiplicity complicates how we typically develop and deploy
machine learning models. We study how multiplicity affects predictions -- i.e.,
predictive multiplicity -- in probabilistic classification. We introduce new
measures for this setting and present optimization-based methods to compute
these measures for convex empirical risk minimization problems like logistic
regression. We apply our methodology to gain insight into why predictive
multiplicity arises. We study the incidence and prevalence of predictive
multiplicity in real-world risk assessment tasks. Our results emphasize the
need to report multiplicity more widely.
|
110,805 | 110,805 | A Communication-efficient Algorithm with Linear Convergence for
Federated Minimax Learning | In this paper, we study a large-scale multi-agent minimax optimization
problem, which models many interesting applications in statistical learning and
game theory, including Generative Adversarial Networks (GANs). The overall
objective is a sum of agents' private local objective functions. We first
analyze an important special case, empirical minimax problem, where the overall
objective approximates a true population minimax risk by statistical samples.
We provide generalization bounds for learning with this objective through
Rademacher complexity analysis. Then, we focus on the federated setting, where
agents can perform local computation and communicate with a central server.
Most existing federated minimax algorithms either require communication per
iteration or lack performance guarantees with the exception of Local Stochastic
Gradient Descent Ascent (SGDA), a multiple-local-update descent ascent
algorithm which guarantees convergence under a diminishing stepsize. By
analyzing Local SGDA under the ideal condition of no gradient noise, we show
that generally it cannot guarantee exact convergence with constant stepsizes
and thus suffers from slow rates of convergence. To tackle this issue, we
propose FedGDA-GT, an improved Federated (Fed) Gradient Descent Ascent (GDA)
method based on Gradient Tracking (GT). When local objectives are Lipschitz
smooth and strongly-convex-strongly-concave, we prove that FedGDA-GT converges
linearly with a constant stepsize to global $\epsilon$-approximation solution
with $\mathcal{O}(\log (1/\epsilon))$ rounds of communication, which matches
the time complexity of centralized GDA method. Finally, we numerically show
that FedGDA-GT outperforms Local SGDA.
|
110,806 | 110,806 | Vygotskian Autotelic Artificial Intelligence: Language and Culture
Internalization for Human-Like AI | Building autonomous artificial agents able to grow open-ended repertoires of
skills is one of the fundamental goals of AI. To that end, a promising
developmental approach recommends the design of intrinsically motivated agents
that learn new skills by generating and pursuing their own goals - autotelic
agents. However, existing algorithms still show serious limitations in terms of
goal diversity, exploration, generalization or skill composition. This
perspective calls for the immersion of autotelic agents into rich
socio-cultural worlds. We focus on language especially, and how its structure
and content may support the development of new cognitive functions in
artificial agents, just like it does in humans. Indeed, most of our skills
could not be learned in isolation. Formal education teaches us to reason
systematically, books teach us history, and YouTube might teach us how to cook.
Crucially, our values, traditions, norms and most of our goals are cultural in
essence. This knowledge, and some argue, some of our cognitive functions such
as abstraction, compositional imagination or relational thinking, are formed
through linguistic and cultural interactions. Inspired by the work of Vygotsky,
we suggest the design of Vygotskian autotelic agents able to interact with
others and, more importantly, able to internalize these interactions to
transform them into cognitive tools supporting the development of new cognitive
functions. This perspective paper proposes a new AI paradigm in the quest for
artificial lifelong skill discovery. It justifies the approach by uncovering
examples of new artificial cognitive functions emerging from interactions
between language and embodiment in recent works at the intersection of deep
reinforcement learning and natural language processing. Looking forward, it
highlights future opportunities and challenges for Vygotskian Autotelic AI
research.
|
110,807 | 110,807 | Finding the Right Recipe for Low Resource Domain Adaptation in Neural
Machine Translation | General translation models often still struggle to generate accurate
translations in specialized domains. To guide machine translation practitioners
and characterize the effectiveness of domain adaptation methods under different
data availability scenarios, we conduct an in-depth empirical exploration of
monolingual and parallel data approaches to domain adaptation of pre-trained,
third-party, NMT models in settings where architecture change is impractical.
We compare data centric adaptation methods in isolation and combination. We
study method effectiveness in very low resource (8k parallel examples) and
moderately low resource (46k parallel examples) conditions and propose an
ensemble approach to alleviate reductions in original domain translation
quality. Our work includes three domains: consumer electronic, clinical, and
biomedical and spans four language pairs - Zh-En, Ja-En, Es-En, and Ru-En. We
also make concrete recommendations for achieving high in-domain performance and
release our consumer electronic and medical domain datasets for all languages
and make our code publicly available.
|
110,808 | 110,808 | Causal Structure Learning: a Combinatorial Perspective | In this review, we discuss approaches for learning causal structure from
data, also called causal discovery. In particular, we focus on approaches for
learning directed acyclic graphs (DAGs) and various generalizations which allow
for some variables to be unobserved in the available data. We devote special
attention to two fundamental combinatorial aspects of causal structure
learning. First, we discuss the structure of the search space over causal
graphs. Second, we discuss the structure of equivalence classes over causal
graphs, i.e., sets of graphs which represent what can be learned from
observational data alone, and how these equivalence classes can be refined by
adding interventional data.
|
110,809 | 110,809 | Posterior Coreset Construction with Kernelized Stein Discrepancy for
Model-Based Reinforcement Learning | In this work, we propose a novel ${\bf K}$ernelized ${\bf S}$tein
Discrepancy-based Posterior Sampling for ${\bf RL}$ algorithm (named
$\texttt{KSRL}$) which extends model-based RL based upon posterior sampling
(PSRL) in several ways: we (i) relax the need for any smoothness or Gaussian
assumptions, allowing for complex mixture models; (ii) ensure it is applicable
to large-scale training by incorporating a compression step such that the
posterior consists of a \emph{Bayesian coreset} of only statistically
significant past state-action pairs; and (iii) develop a novel regret analysis
of PSRL based upon integral probability metrics, which, under a smoothness
condition on the constructed posterior, can be evaluated in closed form as the
kernelized Stein discrepancy (KSD). Consequently, we are able to improve the
$\mathcal{O}(H^{3/2}d\sqrt{T})$ {regret} of PSRL to
$\mathcal{O}(H^{3/2}\sqrt{T})$, where $d$ is the input dimension, $H$ is the
episode length, and $T$ is the total number of episodes experienced,
alleviating a linear dependence on $d$ . Moreover, we theoretically establish a
trade-off between regret rate with posterior representational complexity via
introducing a compression budget parameter $\epsilon$ based on KSD, and
establish a lower bound on the required complexity for consistency of the
model. Experimentally, we observe that this approach is competitive with
several state of the art RL methodologies, with substantive improvements in
computation time. Experimentally, we observe that this approach is competitive
with several state of the art RL methodologies, and can achieve up-to $50\%$
reduction in wall clock time in some continuous control environments.
|
110,810 | 110,810 | Invertible Neural Networks for Graph Prediction | In this work, we address conditional generation using deep invertible neural
networks. This is a type of problem where one aims to infer the most probable
inputs $X$ given outcomes $Y$. We call our method \textit{invertible graph
neural network} (iGNN) due to the primary focus on generating node features on
graph data. A notable feature of our proposed methods is that during network
training, we revise the typically-used loss objective in normalizing flow and
consider Wasserstein-2 regularization to facilitate the training process.
Algorithmic-wise, we adopt an end-to-end training approach since our objective
is to address prediction and generation in the forward and backward processes
at once through a single model. Theoretically, we characterize the conditions
for identifiability of a true mapping, the existence and invertibility of the
mapping, and the expressiveness of iGNN in learning the mapping.
Experimentally, we verify the performance of iGNN on both simulated and
real-data datasets. We demonstrate through extensive numerical experiments that
iGNN shows clear improvement over competing conditional generation benchmarks
on high-dimensional and/or non-convex data.
|
110,811 | 110,811 | Sparse Mixed Linear Regression with Guarantees: Taming an Intractable
Problem with Invex Relaxation | In this paper, we study the problem of sparse mixed linear regression on an
unlabeled dataset that is generated from linear measurements from two different
regression parameter vectors. Since the data is unlabeled, our task is not only
to figure out a good approximation of the regression parameter vectors but also
to label the dataset correctly. In its original form, this problem is NP-hard.
The most popular algorithms to solve this problem (such as
Expectation-Maximization) have a tendency to stuck at local minima. We provide
a novel invex relaxation for this intractable problem which leads to a solution
with provable theoretical guarantees. This relaxation enables exact recovery of
data labels. Furthermore, we recover a close approximation of the regression
parameter vectors which match the true parameter vectors in support and sign.
Our formulation uses a carefully constructed primal dual witnesses framework
for the invex problem. Furthermore, we show that the sample complexity of our
method is only logarithmic in terms of the dimension of the regression
parameter vectors.
|
110,812 | 110,812 | Robust Longitudinal Control for Vehicular Autonomous Platoons Using Deep
Reinforcement Learning | In the last few years, researchers have applied machine learning strategies
in the context of vehicular platoons to increase the safety and efficiency of
cooperative transportation. Reinforcement Learning methods have been employed
in the longitudinal spacing control of Cooperative Adaptive Cruise Control
systems, but to date, none of those studies have addressed problems of
disturbance rejection in such scenarios. Characteristics such as uncertain
parameters in the model and external interferences may prevent agents from
reaching null-spacing errors when traveling at cruising speed. On the other
hand, complex communication topologies lead to specific training processes that
can not be generalized to other contexts, demanding re-training every time the
configuration changes. Therefore, in this paper, we propose an approach to
generalize the training process of a vehicular platoon, such that the
acceleration command of each agent becomes independent of the network topology.
Also, we have modeled the acceleration input as a term with integral action,
such that the Convolutional Neural Network is capable of learning corrective
actions when the states are disturbed by unknown effects. We illustrate the
effectiveness of our proposal with experiments using different network
topologies, uncertain parameters, and external forces. Comparative analyses, in
terms of the steady-state error and overshoot response, were conducted against
the state-of-the-art literature. The findings offer new insights concerning
generalization and robustness of using Reinforcement Learning in the control of
autonomous platoons.
|
110,813 | 110,813 | From Cities to Series: Complex Networks and Deep Learning for Improved
Spatial and Temporal Analytics* | Graphs have often been used to answer questions about the interaction between
real-world entities by taking advantage of their capacity to represent complex
topologies. Complex networks are known to be graphs that capture such
non-trivial topologies; they are able to represent human phenomena such as
epidemic processes, the dynamics of populations, and the urbanization of
cities. The investigation of complex networks has been extrapolated to many
fields of science, with particular emphasis on computing techniques, including
artificial intelligence. In such a case, the analysis of the interaction
between entities of interest is transposed to the internal learning of
algorithms, a paradigm whose investigation is able to expand the state of the
art in Computer Science. By exploring this paradigm, this thesis puts together
complex networks and machine learning techniques to improve the understanding
of the human phenomena observed in pandemics, pendular migration, and street
networks. Accordingly, we contribute with: (i) a new neural network
architecture capable of modeling dynamic processes observed in spatial and
temporal data with applications in epidemics propagation, weather forecasting,
and patient monitoring in intensive care units; (ii) a machine-learning
methodology for analyzing and predicting links in the scope of human mobility
between all the cities of Brazil; and, (iii) techniques for identifying
inconsistencies in the urban planning of cities while tracking the most
influential vertices, with applications over Brazilian and worldwide cities. We
obtained results sustained by sound evidence of advances to the state of the
art in artificial intelligence, rigorous formalisms, and ample experimentation.
Our findings rely upon real-world applications in a range of domains,
demonstrating the applicability of our methodologies.
|
110,814 | 110,814 | Deep Learning on Implicit Neural Datasets | Implicit neural representations (INRs) have become fast, lightweight tools
for storing continuous data, but to date there is no general method for
learning directly with INRs as a data representation. We introduce a principled
deep learning framework for learning and inference directly with INRs of any
type without reverting to grid-based features or operations. Our INR-Nets
evaluate INRs on a low discrepancy sequence, enabling quasi-Monte Carlo (QMC)
integration throughout the network. We prove INR-Nets are universal
approximators on a large class of maps between $L^2$ functions. Additionally,
INR-Nets have convergent gradients under the empirical measure, enabling
backpropagation. We design INR-Nets as a continuous generalization of discrete
networks, enabling them to be initialized with pre-trained models. We
demonstrate learning of INR-Nets on classification (INR$\to$label) and
segmentation (INR$\to$INR) tasks.
|
110,815 | 110,815 | ORC: Network Group-based Knowledge Distillation using Online Role Change | In knowledge distillation, since a single, omnipotent teacher network cannot
solve all problems, multiple teacher-based knowledge distillations have been
studied recently. However, sometimes their improvements are not as good as
expected because some immature teachers may transfer the false knowledge to the
student. In this paper, to overcome this limitation and take the efficacy of
the multiple networks, we divide the multiple networks into teacher and student
groups, respectively. That is, the student group is a set of immature networks
that require learning the teacher's knowledge, while the teacher group consists
of the selected networks that have performed well. Furthermore, according to
our online role change strategy, the top-ranked networks in the student group
are able to promote to the teacher group at every iteration and vice versa.
After training the teacher group using the error images of the student group to
refine the teacher group's knowledge, we transfer the collective knowledge from
the teacher group to the student group successfully. We verify the superiority
of the proposed method on CIFAR-10 and CIFAR-100, which achieves high
performance. We further show the generality of our method with various backbone
architectures such as resent, wrn, vgg, mobilenet, and shufflenet.
|
110,816 | 110,816 | Uniqueness and Complexity of Inverse MDP Models | What is the action sequence aa'a" that was likely responsible for reaching
state s"' (from state s) in 3 steps? Addressing such questions is important in
causal reasoning and in reinforcement learning. Inverse "MDP" models
p(aa'a"|ss"') can be used to answer them. In the traditional "forward" view,
transition "matrix" p(s'|sa) and policy {\pi}(a|s) uniquely determine
"everything": the whole dynamics p(as'a's"a"...|s), and with it, the
action-conditional state process p(s's"...|saa'a"), the multi-step inverse
models p(aa'a"...|ss^i), etc. If the latter is our primary concern, a natural
question, analogous to the forward case is to which extent 1-step inverse model
p(a|ss') plus policy {\pi}(a|s) determine the multi-step inverse models or even
the whole dynamics. In other words, can forward models be inferred from inverse
models or even be side-stepped. This work addresses this question and
variations thereof, and also whether there are efficient decision/inference
algorithms for this.
|
110,817 | 110,817 | Hard Negative Sampling Strategies for Contrastive Representation
Learning | One of the challenges in contrastive learning is the selection of appropriate
\textit{hard negative} examples, in the absence of label information. Random
sampling or importance sampling methods based on feature similarity often lead
to sub-optimal performance. In this work, we introduce UnReMix, a hard negative
sampling strategy that takes into account anchor similarity, model uncertainty
and representativeness. Experimental results on several benchmarks show that
UnReMix improves negative sample selection, and subsequently downstream
performance when compared to state-of-the-art contrastive learning methods.
|
110,818 | 110,818 | Unveiling The Mask of Position-Information Pattern Through the Mist of
Image Features | Recent studies show that paddings in convolutional neural networks encode
absolute position information which can negatively affect the model performance
for certain tasks. However, existing metrics for quantifying the strength of
positional information remain unreliable and frequently lead to erroneous
results. To address this issue, we propose novel metrics for measuring (and
visualizing) the encoded positional information. We formally define the encoded
information as PPP (Position-information Pattern from Padding) and conduct a
series of experiments to study its properties as well as its formation. The
proposed metrics measure the presence of positional information more reliably
than the existing metrics based on PosENet and a test in F-Conv. We also
demonstrate that for any extant (and proposed) padding schemes, PPP is
primarily a learning artifact and is less dependent on the characteristics of
the underlying padding schemes.
|
110,819 | 110,819 | Snow Mountain: Dataset of Audio Recordings of The Bible in Low Resource
Languages | Automatic Speech Recognition (ASR) has increasing utility in the modern
world. There are a many ASR models available for languages with large amounts
of training data like English. However, low-resource languages are poorly
represented. In response we create and release an open-licensed and formatted
dataset of audio recordings of the Bible in low-resource northern Indian
languages. We setup multiple experimental splits and train and analyze two
competitive ASR models to serve as the baseline for future research using this
data.
|
110,820 | 110,820 | Positive Unlabeled Contrastive Learning | Self-supervised pretraining on unlabeled data followed by supervised
finetuning on labeled data is a popular paradigm for learning from limited
labeled examples. In this paper, we investigate and extend this paradigm to the
classical positive unlabeled (PU) setting - the weakly supervised task of
learning a binary classifier only using a few labeled positive examples and a
set of unlabeled samples. We propose a novel PU learning objective positive
unlabeled Noise Contrastive Estimation (puNCE) that leverages the available
explicit (from labeled samples) and implicit (from unlabeled samples)
supervision to learn useful representations from positive unlabeled input data.
The underlying idea is to assign each training sample an individual weight;
labeled positives are given unit weight; unlabeled samples are duplicated, one
copy is labeled positive and the other as negative with weights $\pi$ and
$(1-\pi)$ where $\pi$ denotes the class prior. Extensive experiments across
vision and natural language tasks reveal that puNCE consistently improves over
existing unsupervised and supervised contrastive baselines under limited
supervision.
|
110,821 | 110,821 | RACA: Relation-Aware Credit Assignment for Ad-Hoc Cooperation in
Multi-Agent Deep Reinforcement Learning | In recent years, reinforcement learning has faced several challenges in the
multi-agent domain, such as the credit assignment issue. Value function
factorization emerges as a promising way to handle the credit assignment issue
under the centralized training with decentralized execution (CTDE) paradigm.
However, existing value function factorization methods cannot deal with ad-hoc
cooperation, that is, adapting to new configurations of teammates at test time.
Specifically, these methods do not explicitly utilize the relationship between
agents and cannot adapt to different sizes of inputs. To address these
limitations, we propose a novel method, called Relation-Aware Credit Assignment
(RACA), which achieves zero-shot generalization in ad-hoc cooperation
scenarios. RACA takes advantage of a graph-based relation encoder to encode the
topological structure between agents. Furthermore, RACA utilizes an
attention-based observation abstraction mechanism that can generalize to an
arbitrary number of teammates with a fixed number of parameters. Experiments
demonstrate that our method outperforms baseline methods on the StarCraftII
micromanagement benchmark and ad-hoc cooperation scenarios.
|
110,822 | 110,822 | Accelerated first-order methods for convex optimization with locally
Lipschitz continuous gradient | In this paper we develop accelerated first-order methods for convex
optimization with locally Lipschitz continuous gradient (LLCG), which is beyond
the well-studied class of convex optimization with Lipschitz continuous
gradient. In particular, we first consider unconstrained convex optimization
with LLCG and propose accelerated proximal gradient (APG) methods for solving
it. The proposed APG methods are equipped with a verifiable termination
criterion and enjoy an operation complexity of ${\cal O}(\varepsilon^{-1/2}\log
\varepsilon^{-1})$ and ${\cal O}(\log \varepsilon^{-1})$ for finding an
$\varepsilon$-residual solution of an unconstrained convex and strongly convex
optimization problem, respectively. We then consider constrained convex
optimization with LLCG and propose an first-order proximal augmented Lagrangian
method for solving it by applying one of our proposed APG methods to
approximately solve a sequence of proximal augmented Lagrangian subproblems.
The resulting method is equipped with a verifiable termination criterion and
enjoys an operation complexity of ${\cal O}(\varepsilon^{-1}\log
\varepsilon^{-1})$ and ${\cal O}(\varepsilon^{-1/2}\log \varepsilon^{-1})$ for
finding an $\varepsilon$-KKT solution of a constrained convex and strongly
convex optimization problem, respectively. All the proposed methods in this
paper are parameter-free or almost parameter-free except that the knowledge on
convexity parameter is required. To the best of our knowledge, no prior studies
were conducted to investigate accelerated first-order methods with complexity
guarantees for convex optimization with LLCG. All the complexity results
obtained in this paper are entirely new.
|
110,823 | 110,823 | Equivariant Reinforcement Learning for Quadrotor UAV | This paper presents an equivariant reinforcement learning framework for
quadrotor unmanned aerial vehicles. Successful training of reinforcement
learning often requires numerous interactions with the environments, which
hinders its applicability especially when the available computational resources
are limited, or when there is no reliable simulation model. We identified an
equivariance property of the quadrotor dynamics such that the dimension of the
state required in the training is reduced by one, thereby improving the
sampling efficiency of reinforcement learning substantially. This is
illustrated by numerical examples with popular reinforcement learning
techniques of TD3 and SAC.
|
110,824 | 110,824 | Stochastic gradient descent introduces an effective landscape-dependent
regularization favoring flat solutions | Generalization is one of the most important problems in deep learning (DL).
In the overparameterized regime in neural networks, there exist many low-loss
solutions that fit the training data equally well. The key question is which
solution is more generalizable. Empirical studies showed a strong correlation
between flatness of the loss landscape at a solution and its generalizability,
and stochastic gradient descent (SGD) is crucial in finding the flat solutions.
To understand how SGD drives the learning system to flat solutions, we
construct a simple model whose loss landscape has a continuous set of
degenerate (or near degenerate) minima. By solving the Fokker-Planck equation
of the underlying stochastic learning dynamics, we show that due to its strong
anisotropy the SGD noise introduces an additional effective loss term that
decreases with flatness and has an overall strength that increases with the
learning rate and batch-to-batch variation. We find that the additional
landscape-dependent SGD-loss breaks the degeneracy and serves as an effective
regularization for finding flat solutions. Furthermore, a stronger SGD noise
shortens the convergence time to the flat solutions. However, we identify an
upper bound for the SGD noise beyond which the system fails to converge. Our
results not only elucidate the role of SGD for generalization they may also
have important implications for hyperparameter selection for learning
efficiently without divergence.
|
110,825 | 110,825 | Expressiveness and Learnability: A Unifying View for Evaluating
Self-Supervised Learning | We propose a unifying view to analyze the representation quality of
self-supervised learning (SSL) models without access to supervised labels,
while being agnostic to the architecture, learning algorithm or data
manipulation used during training. We argue that representations can be
evaluated through the lens of expressiveness and learnability. We propose to
use the Intrinsic Dimension (ID) to assess expressiveness and introduce Cluster
Learnability (CL) to assess learnability. CL is measured as the learning speed
of a KNN classifier trained to predict labels obtained by clustering the
representations with K-means. We thus combine CL and ID into a single
predictor: CLID. Through a large-scale empirical study with a diverse family of
SSL algorithms, we find that CLID better correlates with in-distribution model
performance than other competing recent evaluation schemes. We also benchmark
CLID on out-of-domain generalization, where CLID serves as a predictor of the
transfer performance of SSL models on several classification tasks, yielding
improvements with respect to the competing baselines.
|
110,826 | 110,826 | Which Explanation Should I Choose? A Function Approximation Perspective
to Characterizing Post hoc Explanations | Despite the plethora of post hoc model explanation methods, the basic
properties and behavior of these methods and the conditions under which each
one is effective are not well understood. In this work, we bridge these gaps
and address a fundamental question: Which explanation method should one use in
a given situation? To this end, we adopt a function approximation perspective
and formalize the local function approximation (LFA) framework. We show that
popular explanation methods are instances of this framework, performing
function approximations of the underlying model in different neighborhoods
using different loss functions. We introduce a no free lunch theorem for
explanation methods which demonstrates that no single method can perform
optimally across all neighbourhoods and calls for choosing among methods. To
choose among methods, we set forth a guiding principle based on the function
approximation perspective, considering a method to be effective if it recovers
the underlying model when the model is a member of the explanation function
class. Then, we analyze the conditions under which popular explanation methods
are effective and provide recommendations for choosing among explanation
methods and creating new ones. Lastly, we empirically validate our theoretical
results using various real world datasets, model classes, and prediction tasks.
By providing a principled mathematical framework which unifies diverse
explanation methods, our work characterizes the behaviour of these methods and
their relation to one another, guides the choice of explanation methods, and
paves the way for the creation of new ones.
|
110,827 | 110,827 | Compressive Fourier collocation methods for high-dimensional diffusion
equations with periodic boundary conditions | High-dimensional Partial Differential Equations (PDEs) are a popular
mathematical modelling tool, with applications ranging from finance to
computational chemistry. However, standard numerical techniques for solving
these PDEs are typically affected by the curse of dimensionality. In this work,
we tackle this challenge while focusing on stationary diffusion equations
defined over a high-dimensional domain with periodic boundary conditions.
Inspired by recent progress in sparse function approximation in high
dimensions, we propose a new method called compressive Fourier collocation.
Combining ideas from compressive sensing and spectral collocation, our method
replaces the use of structured collocation grids with Monte Carlo sampling and
employs sparse recovery techniques, such as orthogonal matching pursuit and
$\ell^1$ minimization, to approximate the Fourier coefficients of the PDE
solution. We conduct a rigorous theoretical analysis showing that the
approximation error of the proposed method is comparable with the best $s$-term
approximation (with respect to the Fourier basis) to the solution. Using the
recently introduced framework of random sampling in bounded Riesz systems, our
analysis shows that the compressive Fourier collocation method mitigates the
curse of dimensionality with respect to the number of collocation points under
sufficient conditions on the regularity of the diffusion coefficient. We also
present numerical experiments that illustrate the accuracy and stability of the
method for the approximation of sparse and compressible solutions.
|
110,828 | 110,828 | Entangled Residual Mappings | Residual mappings have been shown to perform representation learning in the
first layers and iterative feature refinement in higher layers. This interplay,
combined with their stabilizing effect on the gradient norms, enables them to
train very deep networks. In this paper, we take a step further and introduce
entangled residual mappings to generalize the structure of the residual
connections and evaluate their role in iterative learning representations. An
entangled residual mapping replaces the identity skip connections with
specialized entangled mappings such as orthogonal, sparse, and structural
correlation matrices that share key attributes (eigenvalues, structure, and
Jacobian norm) with identity mappings. We show that while entangled mappings
can preserve the iterative refinement of features across various deep models,
they influence the representation learning process in convolutional networks
differently than attention-based models and recurrent neural networks. In
general, we find that for CNNs and Vision Transformers entangled sparse mapping
can help generalization while orthogonal mappings hurt performance. For
recurrent networks, orthogonal residual mappings form an inductive bias for
time-variant sequences, which degrades accuracy on time-invariant tasks.
|
110,829 | 110,829 | Deep Learning Architecture Based Approach For 2D-Simulation of Microwave
Plasma Interaction | This paper presents a convolutional neural network (CNN)-based deep learning
model, inspired from UNet with series of encoder and decoder units with skip
connections, for the simulation of microwave-plasma interaction. The microwave
propagation characteristics in complex plasma medium pertaining to
transmission, absorption and reflection primarily depends on the ratio of
electromagnetic (EM) wave frequency and electron plasma frequency, and the
plasma density profile. The scattering of a plane EM wave with fixed frequency
(1 GHz) and amplitude incident on a plasma medium with different gaussian
density profiles (in the range of $1\times 10^{17}-1\times 10^{22}{m^{-3}}$)
have been considered. The training data associated with microwave-plasma
interaction has been generated using 2D-FDTD (Finite Difference Time Domain)
based simulations. The trained deep learning model is then used to reproduce
the scattered electric field values for the 1GHz incident microwave on
different plasma profiles with error margin of less than 2\%. We propose a
complete deep learning (DL) based pipeline to train, validate and evaluate the
model. We compare the results of the network, using various metrics like SSIM
index, average percent error and mean square error, with the physical data
obtained from well-established FDTD based EM solvers. To the best of our
knowledge, this is the first effort towards exploring a DL based approach for
the simulation of complex microwave plasma interaction. The deep learning
technique proposed in this work is significantly fast as compared to the
existing computational techniques, and can be used as a new, prospective and
alternative computational approach for investigating microwave-plasma
interaction in a real time scenario.
|
110,830 | 110,830 | Exponential Separations in Symmetric Neural Networks | In this work we demonstrate a novel separation between symmetric neural
network architectures. Specifically, we consider the Relational
Network~\parencite{santoro2017simple} architecture as a natural generalization
of the DeepSets~\parencite{zaheer2017deep} architecture, and study their
representational gap. Under the restriction to analytic activation functions,
we construct a symmetric function acting on sets of size $N$ with elements in
dimension $D$, which can be efficiently approximated by the former
architecture, but provably requires width exponential in $N$ and $D$ for the
latter.
|
110,831 | 110,831 | Algorithmic Stability of Heavy-Tailed Stochastic Gradient Descent on
Least Squares | Recent studies have shown that heavy tails can emerge in stochastic
optimization and that the heaviness of the tails has links to the
generalization error. While these studies have shed light on interesting
aspects of the generalization behavior in modern settings, they relied on
strong topological and statistical regularity assumptions, which are hard to
verify in practice. Furthermore, it has been empirically illustrated that the
relation between heavy tails and generalization might not always be monotonic
in practice, contrary to the conclusions of existing theory. In this study, we
establish novel links between the tail behavior and generalization properties
of stochastic gradient descent (SGD), through the lens of algorithmic
stability. We consider a quadratic optimization problem and use a heavy-tailed
stochastic differential equation as a proxy for modeling the heavy-tailed
behavior emerging in SGD. We then prove uniform stability bounds, which reveal
the following outcomes: (i) Without making any exotic assumptions, we show that
SGD will not be stable if the stability is measured with the squared-loss
$x\mapsto x^2$, whereas it in turn becomes stable if the stability is instead
measured with a surrogate loss $x\mapsto |x|^p$ with some $p<2$. (ii) Depending
on the variance of the data, there exists a \emph{`threshold of
heavy-tailedness'} such that the generalization error decreases as the tails
become heavier, as long as the tails are lighter than this threshold. This
suggests that the relation between heavy tails and generalization is not
globally monotonic. (iii) We prove matching lower-bounds on uniform stability,
implying that our bounds are tight in terms of the heaviness of the tails. We
support our theory with synthetic and real neural network experiments.
|
110,832 | 110,832 | Lottery Tickets on a Data Diet: Finding Initializations with Sparse
Trainable Networks | A striking observation about iterative magnitude pruning (IMP; Frankle et al.
2020) is that $\unicode{x2014}$ after just a few hundred steps of dense
training $\unicode{x2014}$ the method can find a sparse sub-network that can be
trained to the same accuracy as the dense network. However, the same does not
hold at step 0, i.e. random initialization. In this work, we seek to understand
how this early phase of pre-training leads to a good initialization for IMP
both through the lens of the data distribution and the loss landscape geometry.
Empirically we observe that, holding the number of pre-training iterations
constant, training on a small fraction of (randomly chosen) data suffices to
obtain an equally good initialization for IMP. We additionally observe that by
pre-training only on "easy" training data, we can decrease the number of steps
necessary to find a good initialization for IMP compared to training on the
full dataset or a randomly chosen subset. Finally, we identify novel properties
of the loss landscape of dense networks that are predictive of IMP performance,
showing in particular that more examples being linearly mode connected in the
dense network correlates well with good initializations for IMP. Combined,
these results provide new insight into the role played by the early phase
training in IMP.
|
110,833 | 110,833 | Sequential Permutation Testing of Random Forest Variable Importance
Measures | Hypothesis testing of random forest (RF) variable importance measures (VIMP)
remains the subject of ongoing research. Among recent developments, heuristic
approaches to parametric testing have been proposed whose distributional
assumptions are based on empirical evidence. Other formal tests under
regularity conditions were derived analytically. However, these approaches can
be computationally expensive or even practically infeasible. This problem also
occurs with non-parametric permutation tests, which are, however,
distribution-free and can generically be applied to any type of RF and VIMP.
Embracing this advantage, it is proposed here to use sequential permutation
tests and sequential p-value estimation to reduce the high computational costs
associated with conventional permutation tests. The popular and widely used
permutation VIMP serves as a practical and relevant application example. The
results of simulation studies confirm that the theoretical properties of the
sequential tests apply, that is, the type-I error probability is controlled at
a nominal level and a high power is maintained with considerably fewer
permutations needed in comparison to conventional permutation testing. The
numerical stability of the methods is investigated in two additional
application studies. In summary, theoretically sound sequential permutation
testing of VIMP is possible at greatly reduced computational costs.
Recommendations for application are given. A respective implementation is
provided through the accompanying R package $rfvimptest$. The approach can also
be easily applied to any kind of prediction model.
|
110,834 | 110,834 | Decentralized Training of Foundation Models in Heterogeneous
Environments | Training foundation models, such as GPT-3 and PaLM, can be extremely
expensive, often involving tens of thousands of GPUs running continuously for
months. These models are typically trained in specialized clusters featuring
fast, homogeneous interconnects and using carefully designed software systems
that support both data parallelism and model/pipeline parallelism. Such
dedicated clusters can be costly and difficult to obtain. Can we instead
leverage the much greater amount of decentralized, heterogeneous, and
lower-bandwidth interconnected compute? Previous works examining the
heterogeneous, decentralized setting focus on relatively small models that can
be trained in a purely data parallel manner. State-of-the-art schemes for model
parallel foundation model training, such as Megatron, only consider the
homogeneous data center setting. In this paper, we present the first study of
training large foundation models with model parallelism in a decentralized
regime over a heterogeneous network. Our key technical contribution is a
scheduling algorithm that allocates different computational "tasklets" in the
training of foundation models to a group of decentralized GPU devices connected
by a slow heterogeneous network. We provide a formal cost model and further
propose an efficient evolutionary algorithm to find the optimal allocation
strategy. We conduct extensive experiments that represent different scenarios
for learning over geo-distributed devices simulated using real-world network
measurements. In the most extreme case, across 8 different cities spanning 3
continents, our approach is 4.8X faster than prior state-of-the-art training
systems (Megatron).
|
110,835 | 110,835 | Incrementality Bidding via Reinforcement Learning under Mixed and
Delayed Rewards | Incrementality, which is used to measure the causal effect of showing an ad
to a potential customer (e.g. a user in an internet platform) versus not, is a
central object for advertisers in online advertising platforms. This paper
investigates the problem of how an advertiser can learn to optimize the bidding
sequence in an online manner \emph{without} knowing the incrementality
parameters in advance. We formulate the offline version of this problem as a
specially structured episodic Markov Decision Process (MDP) and then, for its
online learning counterpart, propose a novel reinforcement learning (RL)
algorithm with regret at most $\widetilde{O}(H^2\sqrt{T})$, which depends on
the number of rounds $H$ and number of episodes $T$, but does not depend on the
number of actions (i.e., possible bids). A fundamental difference between our
learning problem from standard RL problems is that the realized reward feedback
from conversion incrementality is \emph{mixed} and \emph{delayed}. To handle
this difficulty we propose and analyze a novel pairwise moment-matching
algorithm to learn the conversion incrementality, which we believe is of
independent of interest.
|
110,836 | 110,836 | Rashomon Capacity: A Metric for Predictive Multiplicity in Probabilistic
Classification | Predictive multiplicity occurs when classification models with nearly
indistinguishable average performances assign conflicting predictions to
individual samples. When used for decision-making in applications of
consequence (e.g., lending, education, criminal justice), models developed
without regard for predictive multiplicity may result in unjustified and
arbitrary decisions for specific individuals. We introduce a new measure of
predictive multiplicity in probabilistic classification called Rashomon
Capacity. Prior metrics for predictive multiplicity focus on classifiers that
output thresholded (i.e., 0-1) predicted classes. In contrast, Rashomon
Capacity applies to probabilistic classifiers, capturing more nuanced score
variations for individual samples. We provide a rigorous derivation for
Rashomon Capacity, argue its intuitive appeal, and demonstrate how to estimate
it in practice. We show that Rashomon Capacity yields principled strategies for
disclosing conflicting models to stakeholders. Our numerical experiments
illustrate how Rashomon Capacity captures predictive multiplicity in various
datasets and learning models, including neural networks. The tools introduced
in this paper can help data scientists measure, report, and ultimately resolve
predictive multiplicity prior to model deployment.
|
110,837 | 110,837 | PNODE: A memory-efficient neural ODE framework based on high-level
adjoint differentiation | Neural ordinary differential equations (neural ODEs) have emerged as a novel
network architecture that bridges dynamical systems and deep learning. However,
the gradient obtained with the continuous adjoint method in the vanilla neural
ODE is not reverse-accurate. Other approaches suffer either from excessive
memory requirement due to deep computational graphs or from limited choices for
the time integration scheme, hampering their application to large-scale complex
dynamical systems. To achieve accurate gradients without compromising memory
efficiency and flexibility, we present a new neural ODE framework, PNODE, based
on high-level discrete adjoint algorithmic differentiation. By leveraging
discrete adjoint time integrators and advanced checkpointing strategies
tailored for these integrators, PNODE can provide a balance between memory and
computational costs, while computing the gradients consistently and accurately.
We provide an open-source implementation based on PyTorch and PETSc, one of the
most commonly used portable, scalable scientific computing libraries. We
demonstrate the performance through extensive numerical experiments on image
classification and continuous normalizing flow problems. We show that PNODE
achieves the highest memory efficiency when compared with other
reverse-accurate methods. On the image classification problems, PNODE is up to
two times faster than the vanilla neural ODE and up to 2.3 times faster than
the best existing reverse-accurate method. We also show that PNODE enables the
use of the implicit time integration methods that are needed for stiff
dynamical systems.
|
110,838 | 110,838 | Fine-tuning Language Models over Slow Networks using Activation
Compression with Guarantees | Communication compression is a crucial technique for modern distributed
learning systems to alleviate their communication bottlenecks over slower
networks. Despite recent intensive studies of gradient compression for data
parallel-style training, compressing the activations for models trained with
pipeline parallelism is still an open problem. In this paper, we propose
AC-SGD, a novel activation compression algorithm for communication-efficient
pipeline parallelism training over slow networks. Different from previous
efforts in activation compression, instead of compressing activation values
directly, AC-SGD compresses the changes of the activations. This allows us to
show, to the best of our knowledge for the first time, that one can still
achieve $O(1/\sqrt{T})$ convergence rate for non-convex objectives under
activation compression, without making assumptions on gradient unbiasedness
that do not hold for deep learning models with non-linear activation
functions.We then show that AC-SGD can be optimized and implemented
efficiently, without additional end-to-end runtime overhead.We evaluated AC-SGD
to fine-tune language models with up to 1.5 billion parameters, compressing
activations to 2-4 bits.AC-SGD provides up to 4.3X end-to-end speed-up in
slower networks, without sacrificing model quality. Moreover, we also show that
AC-SGD can be combined with state-of-the-art gradient compression algorithms to
enable "end-to-end communication compression: All communications between
machines, including model gradients, forward activations, and backward
gradients are compressed into lower precision.This provides up to 4.9X
end-to-end speed-up, without sacrificing model quality.
|
110,839 | 110,839 | Learning a Restricted Boltzmann Machine using biased Monte Carlo
sampling | Restricted Boltzmann Machines are simple and powerful generative models
capable of encoding any complex dataset. Despite all their advantages, in
practice, trainings are often unstable, and it is hard to assess their quality
because dynamics are hampered by extremely slow time-dependencies. This
situation becomes critical when dealing with low-dimensional clustered
datasets, where the time needed to sample ergodically the trained models
becomes computationally prohibitive. In this work, we show that this divergence
of Monte Carlo mixing times is related to a phase coexistence phenomenon,
similar to that encountered in Physics in the vicinity of a first order phase
transition. We show that sampling the equilibrium distribution via Markov Chain
Monte Carlo can be dramatically accelerated using biased sampling techniques,
in particular, the Tethered Monte Carlo method (TMC). This sampling technique
solves efficiently the problem of evaluating the quality of a given trained
model and the generation of new samples in reasonable times. In addition, we
show that this sampling technique can be exploited to improve the computation
of the log-likelihood gradient during the training too, which produces dramatic
improvements when training RBMs with artificial clustered datasets. When
dealing with real low-dimensional datasets, this new training procedure fits
RBM models with significantly faster relaxational dynamics than those obtained
with standard PCD recipes. We also show that TMC sampling can be used to
recover free-energy profile of the RBM, which turns out to be extremely useful
to compute the probability distribution of a given model and to improve the
generation of new decorrelated samples on slow PCD trained models.
|
110,840 | 110,840 | Learning Soft Constraints From Constrained Expert Demonstrations | Inverse reinforcement learning (IRL) methods assume that the expert data is
generated by an agent optimizing some reward function. However, in many
settings, the agent may optimize a reward function subject to some constraints,
where the constraints induce behaviors that may be otherwise difficult to
express with just a reward function. We consider the setting where the reward
function is given, and the constraints are unknown, and propose a method that
is able to recover these constraints satisfactorily from the expert data. While
previous work has focused on recovering hard constraints, our method can
recover cumulative soft constraints that the agent satisfies on average per
episode. In IRL fashion, our method solves this problem by adjusting the
constraint function iteratively through a constrained optimization procedure,
until the agent behavior matches the expert behavior. Despite the simplicity of
the formulation, our method is able to obtain good results. We demonstrate our
approach on synthetic environments and real world highway driving data.
|
110,841 | 110,841 | A New Security Boundary of Component Differentially Challenged XOR PUFs
Against Machine Learning Modeling Attacks | Physical Unclonable Functions (PUFs) are promising security primitives for
resource-constrained network nodes. The XOR Arbiter PUF (XOR PUF or XPUF) is an
intensively studied PUF invented to improve the security of the Arbiter PUF,
probably the most lightweight delay-based PUF. Recently, highly powerful
machine learning attack methods were discovered and were able to easily break
large-sized XPUFs, which were highly secure against earlier machine learning
attack methods. Component-differentially-challenged XPUFs (CDC-XPUFs) are XPUFs
with different component PUFs receiving different challenges. Studies showed
they were much more secure against machine learning attacks than the
conventional XPUFs, whose component PUFs receive the same challenge. But these
studies were all based on earlier machine learning attack methods, and hence it
is not clear if CDC-XPUFs can remain secure under the recently discovered
powerful attack methods. In this paper, the two current most powerful two
machine learning methods for attacking XPUFs are adapted by fine-tuning the
parameters of the two methods for CDC-XPUFs. Attack experiments using both
simulated PUF data and silicon data generated from PUFs implemented on
field-programmable gate array (FPGA) were carried out, and the experimental
results showed that some previously secure CDC-XPUFs of certain circuit
parameter values are no longer secure under the adapted new attack methods,
while many more CDC-XPUFs of other circuit parameter values remain secure.
Thus, our experimental attack study has re-defined the boundary between the
secure region and the insecure region of the PUF circuit parameter space,
providing PUF manufacturers and IoT security application developers with
valuable information in choosing PUFs with secure parameter values.
|
110,842 | 110,842 | Sample-Efficient Reinforcement Learning of Partially Observable Markov
Games | This paper considers the challenging tasks of Multi-Agent Reinforcement
Learning (MARL) under partial observability, where each agent only sees her own
individual observations and actions that reveal incomplete information about
the underlying state of system. This paper studies these tasks under the
general model of multiplayer general-sum Partially Observable Markov Games
(POMGs), which is significantly larger than the standard model of Imperfect
Information Extensive-Form Games (IIEFGs). We identify a rich subclass of POMGs
-- weakly revealing POMGs -- in which sample-efficient learning is tractable.
In the self-play setting, we prove that a simple algorithm combining optimism
and Maximum Likelihood Estimation (MLE) is sufficient to find approximate Nash
equilibria, correlated equilibria, as well as coarse correlated equilibria of
weakly revealing POMGs, in a polynomial number of samples when the number of
agents is small. In the setting of playing against adversarial opponents, we
show that a variant of our optimistic MLE algorithm is capable of achieving
sublinear regret when being compared against the optimal maximin policies. To
our best knowledge, this work provides the first line of sample-efficient
results for learning POMGs.
|
110,843 | 110,843 | SPD domain-specific batch normalization to crack interpretable
unsupervised domain adaptation in EEG | Electroencephalography (EEG) provides access to neuronal dynamics
non-invasively with millisecond resolution, rendering it a viable method in
neuroscience and healthcare. However, its utility is limited as current EEG
technology does not generalize well across domains (i.e., sessions and
subjects) without expensive supervised re-calibration. Contemporary methods
cast this transfer learning (TL) problem as a multi-source/-target unsupervised
domain adaptation (UDA) problem and address it with deep learning or shallow,
Riemannian geometry aware alignment methods. Both directions have, so far,
failed to consistently close the performance gap to state-of-the-art
domain-specific methods based on tangent space mapping (TSM) on the symmetric
positive definite (SPD) manifold. Here, we propose a theory-based machine
learning framework that enables, for the first time, learning domain-invariant
TSM models in an end-to-end fashion. To achieve this, we propose a new building
block for geometric deep learning, which we denote SPD domain-specific momentum
batch normalization (SPDDSMBN). A SPDDSMBN layer can transform domain-specific
SPD inputs into domain-invariant SPD outputs, and can be readily applied to
multi-source/-target and online UDA scenarios. In extensive experiments with 6
diverse EEG brain-computer interface (BCI) datasets, we obtain state-of-the-art
performance in inter-session and -subject TL with a simple, intrinsically
interpretable network architecture, which we denote TSMNet.
|
110,844 | 110,844 | Improving Fairness in Large-Scale Object Recognition by CrowdSourced
Demographic Information | There has been increasing awareness of ethical issues in machine learning,
and fairness has become an important research topic. Most fairness efforts in
computer vision have been focused on human sensing applications and preventing
discrimination by people's physical attributes such as race, skin color or age
by increasing visual representation for particular demographic groups. We argue
that ML fairness efforts should extend to object recognition as well.
Buildings, artwork, food and clothing are examples of the objects that define
human culture. Representing these objects fairly in machine learning datasets
will lead to models that are less biased towards a particular culture and more
inclusive of different traditions and values. There exist many research
datasets for object recognition, but they have not carefully considered which
classes should be included, or how much training data should be collected per
class. To address this, we propose a simple and general approach, based on
crowdsourcing the demographic composition of the contributors: we define fair
relevance scores, estimate them, and assign them to each class. We showcase its
application to the landmark recognition domain, presenting a detailed analysis
and the final fairer landmark rankings. We present analysis which leads to a
much fairer coverage of the world compared to existing datasets. The evaluation
dataset was used for the 2021 Google Landmark Challenges, which was the first
of a kind with an emphasis on fairness in generic object recognition.
|
110,845 | 110,845 | Optimal Activation Functions for the Random Features Regression Model | The asymptotic mean squared test error and sensitivity of the Random Features
Regression model (RFR) have been recently studied. We build on this work and
identify in closed-form the family of Activation Functions (AFs) that minimize
a combination of the test error and sensitivity of the RFR under different
notions of functional parsimony. We find scenarios under which the optimal AFs
are linear, saturated linear functions, or expressible in terms of Hermite
polynomials. Finally, we show how using optimal AFs impacts well-established
properties of the RFR model, such as its double descent curve, and the
dependency of its optimal regularization parameter on the observation noise
level.
|
110,846 | 110,846 | Code Generation Tools (Almost) for Free? A Study of Few-Shot,
Pre-Trained Language Models on Code | Few-shot learning with large-scale, pre-trained language models is a powerful
way to answer questions about code, e.g., how to complete a given code example,
or even generate code snippets from scratch. The success of these models raises
the question whether they could serve as a basis for building a wide range code
generation tools. Traditionally, such tools are built manually and separately
for each task. Instead, few-shot learning may allow to obtain different tools
from a single pre-trained language model by simply providing a few examples or
a natural language description of the expected tool behavior. This paper
studies to what extent a state-of-the-art, pre-trained language model of code,
Codex, may serve this purpose. We consider three code manipulation and code
generation tasks targeted by a range of traditional tools: (i) code mutation;
(ii) test oracle generation from natural language documentation; and (iii) test
case generation. For each task, we compare few-shot learning to a manually
built tool. Our results show that the model-based tools complement (code
mutation), are on par (test oracle generation), or even outperform their
respective traditionally built tool (test case generation), while imposing far
less effort to develop them. By comparing the effectiveness of different
variants of the model-based tools, we provide insights on how to design an
appropriate input ("prompt") to the model and what influence the size of the
model has. For example, we find that providing a small natural language
description of the code generation task is an easy way to improve predictions.
Overall, we conclude that few-shot language models are surprisingly effective,
yet there is still more work to be done, such as exploring more diverse ways of
prompting and tackling even more involved tasks.
|
110,847 | 110,847 | Equipping Black-Box Policies with Model-Based Advice for Stable
Nonlinear Control | Machine-learned black-box policies are ubiquitous for nonlinear control
problems. Meanwhile, crude model information is often available for these
problems from, e.g., linear approximations of nonlinear dynamics. We study the
problem of equipping a black-box control policy with model-based advice for
nonlinear control on a single trajectory. We first show a general negative
result that a naive convex combination of a black-box policy and a linear
model-based policy can lead to instability, even if the two policies are both
stabilizing. We then propose an adaptive $\lambda$-confident policy, with a
coefficient $\lambda$ indicating the confidence in a black-box policy, and
prove its stability. With bounded nonlinearity, in addition, we show that the
adaptive $\lambda$-confident policy achieves a bounded competitive ratio when a
black-box policy is near-optimal. Finally, we propose an online learning
approach to implement the adaptive $\lambda$-confident policy and verify its
efficacy in case studies about the CartPole problem and a real-world electric
vehicle (EV) charging problem with data bias due to COVID-19.
|
110,848 | 110,848 | Understanding the Role of Nonlinearity in Training Dynamics of
Contrastive Learning | While the empirical success of self-supervised learning (SSL) heavily relies
on the usage of deep nonlinear models, many theoretical works proposed to
understand SSL still focus on linear ones. In this paper, we study the role of
nonlinearity in the training dynamics of contrastive learning (CL) on one and
two-layer nonlinear networks with homogeneous activation $h(x) = h'(x)x$. We
theoretically demonstrate that (1) the presence of nonlinearity leads to many
local optima even in 1-layer setting, each corresponding to certain patterns
from the data distribution, while with linear activation, only one major
pattern can be learned; and (2) nonlinearity leads to specialized weights into
diverse patterns, a behavior that linear activation is proven not capable of.
These findings suggest that models with lots of parameters can be regarded as a
\emph{brute-force} way to find these local optima induced by nonlinearity, a
possible underlying reason why empirical observations such as the lottery
ticket hypothesis hold. In addition, for 2-layer setting, we also discover
\emph{global modulation}: those local patterns discriminative from the
perspective of global-level patterns are prioritized to learn, further
characterizing the learning process. Simulation verifies our theoretical
findings.
|
110,849 | 110,849 | HEX: Human-in-the-loop Explainability via Deep Reinforcement Learning | The use of machine learning (ML) models in decision-making contexts,
particularly those used in high-stakes decision-making, are fraught with issue
and peril since a person - not a machine - must ultimately be held accountable
for the consequences of the decisions made using such systems. Machine learning
explainability (MLX) promises to provide decision-makers with
prediction-specific rationale, assuring them that the model-elicited
predictions are made for the right reasons and are thus reliable. Few works
explicitly consider this key human-in-the-loop (HITL) component, however. In
this work we propose HEX, a human-in-the-loop deep reinforcement learning
approach to MLX. HEX incorporates 0-distrust projection to synthesize decider
specific explanation-providing policies from any arbitrary classification
model. HEX is also constructed to operate in limited or reduced training data
scenarios, such as those employing federated learning. Our formulation
explicitly considers the decision boundary of the ML model in question, rather
than the underlying training data, which is a shortcoming of many
model-agnostic MLX methods. Our proposed methods thus synthesize HITL MLX
policies that explicitly capture the decision boundary of the model in question
for use in limited data scenarios.
|
110,850 | 110,850 | Detecting Pulmonary Embolism from Computed Tomography Using
Convolutional Neural Network | The clinical symptoms of pulmonary embolism (PE) are very diverse and
non-specific, which makes it difficult to diagnose. In addition, pulmonary
embolism has multiple triggers and is one of the major causes of vascular
death. Therefore, if it can be detected and treated quickly, it can
significantly reduce the risk of death in hospitalized patients. In the
detection process, the cost of computed tomography pulmonary angiography (CTPA)
is high, and angiography requires the injection of contrast agents, which
increase the risk of damage to the patient. Therefore, this study will use a
deep learning approach to detect pulmonary embolism in all patients who take a
CT image of the chest using a convolutional neural network. With the proposed
pulmonary embolism detection system, we can detect the possibility of pulmonary
embolism at the same time as the patient's first CT image, and schedule the
CTPA test immediately, saving more than a week of CT image screening time and
providing timely diagnosis and treatment to the patient.
|
110,851 | 110,851 | MultiHiertt: Numerical Reasoning over Multi Hierarchical Tabular and
Textual Data | Numerical reasoning over hybrid data containing both textual and tabular
content (e.g., financial reports) has recently attracted much attention in the
NLP community. However, existing question answering (QA) benchmarks over hybrid
data only include a single flat table in each document and thus lack examples
of multi-step numerical reasoning across multiple hierarchical tables. To
facilitate data analytical progress, we construct a new large-scale benchmark,
MultiHiertt, with QA pairs over Multi Hierarchical Tabular and Textual data.
MultiHiertt is built from a wealth of financial reports and has the following
unique characteristics: 1) each document contain multiple tables and longer
unstructured texts; 2) most of tables contained are hierarchical; 3) the
reasoning process required for each question is more complex and challenging
than existing benchmarks; and 4) fine-grained annotations of reasoning
processes and supporting facts are provided to reveal complex numerical
reasoning. We further introduce a novel QA model termed MT2Net, which first
applies facts retrieving to extract relevant supporting facts from both tables
and text and then uses a reasoning module to perform symbolic reasoning over
retrieved facts. We conduct comprehensive experiments on various baselines. The
experimental results show that MultiHiertt presents a strong challenge for
existing baselines whose results lag far behind the performance of human
experts. The dataset and code are publicly available at
https://github.com/psunlpgroup/MultiHiertt.
|
110,852 | 110,852 | On the Privacy Properties of GAN-generated Samples | The privacy implications of generative adversarial networks (GANs) are a
topic of great interest, leading to several recent algorithms for training GANs
with privacy guarantees. By drawing connections to the generalization
properties of GANs, we prove that under some assumptions, GAN-generated samples
inherently satisfy some (weak) privacy guarantees. First, we show that if a GAN
is trained on m samples and used to generate n samples, the generated samples
are (epsilon, delta)-differentially-private for (epsilon, delta) pairs where
delta scales as O(n/m). We show that under some special conditions, this upper
bound is tight. Next, we study the robustness of GAN-generated samples to
membership inference attacks. We model membership inference as a hypothesis
test in which the adversary must determine whether a given sample was drawn
from the training dataset or from the underlying data distribution. We show
that this adversary can achieve an area under the ROC curve that scales no
better than O(m^{-1/4}).
|
110,853 | 110,853 | Supernet Training for Federated Image Classification under System
Heterogeneity | Efficient deployment of deep neural networks across many devices and resource
constraints, especially on edge devices, is one of the most challenging
problems in the presence of data-privacy preservation issues. Conventional
approaches have evolved to either improve a single global model while keeping
each local training data decentralized (i.e., data-heterogeneity) or to train a
once-for-all network that supports diverse architectural settings to address
heterogeneous systems equipped with different computational capabilities (i.e.,
model-heterogeneity). However, little research has considered both directions
simultaneously. In this work, we propose a novel framework to consider both
scenarios, namely Federation of Supernet Training (FedSup), where clients send
and receive a supernet whereby it contains all possible architectures sampled
from itself. It is inspired by how averaging parameters in the model
aggregation stage of Federated Learning (FL) is similar to weight-sharing in
supernet training. Specifically, in the FedSup framework, a weight-sharing
approach widely used in the training single shot model is combined with the
averaging of Federated Learning (FedAvg). Under our framework, we present an
efficient algorithm (E-FedSup) by sending the sub-model to clients in the
broadcast stage for reducing communication costs and training overhead. We
demonstrate several strategies to enhance supernet training in the FL
environment and conduct extensive empirical evaluations. The resulting
framework is shown to pave the way for the robustness of both data- and
model-heterogeneity on several standard benchmarks.
|
110,854 | 110,854 | Adversarial Unlearning: Reducing Confidence Along Adversarial Directions | Supervised learning methods trained with maximum likelihood objectives often
overfit on training data. Most regularizers that prevent overfitting look to
increase confidence on additional examples (e.g., data augmentation,
adversarial training), or reduce it on training data (e.g., label smoothing).
In this work we propose a complementary regularization strategy that reduces
confidence on self-generated examples. The method, which we call RCAD (Reducing
Confidence along Adversarial Directions), aims to reduce confidence on
out-of-distribution examples lying along directions adversarially chosen to
increase training loss. In contrast to adversarial training, RCAD does not try
to robustify the model to output the original label, but rather regularizes it
to have reduced confidence on points generated using much larger perturbations
than in conventional adversarial training. RCAD can be easily integrated into
training pipelines with a few lines of code. Despite its simplicity, we find on
many classification benchmarks that RCAD can be added to existing techniques
(e.g., label smoothing, MixUp training) to increase test accuracy by 1-3% in
absolute value, with more significant gains in the low data regime. We also
provide a theoretical analysis that helps to explain these benefits in
simplified settings, showing that RCAD can provably help the model unlearn
spurious features in the training data.
|
110,855 | 110,855 | Incremental Learning Meets Transfer Learning: Application to Multi-site
Prostate MRI Segmentation | Many medical datasets have recently been created for medical image
segmentation tasks, and it is natural to question whether we can use them to
sequentially train a single model that (1) performs better on all these
datasets, and (2) generalizes well and transfers better to the unknown target
site domain. Prior works have achieved this goal by jointly training one model
on multi-site datasets, which achieve competitive performance on average but
such methods rely on the assumption about the availability of all training
data, thus limiting its effectiveness in practical deployment. In this paper,
we propose a novel multi-site segmentation framework called
incremental-transfer learning (ITL), which learns a model from multi-site
datasets in an end-to-end sequential fashion. Specifically, "incremental"
refers to training sequentially constructed datasets, and "transfer" is
achieved by leveraging useful information from the linear combination of
embedding features on each dataset. In addition, we introduce our ITL
framework, where we train the network including a site-agnostic encoder with
pre-trained weights and at most two segmentation decoder heads. We also design
a novel site-level incremental loss in order to generalize well on the target
domain. Second, we show for the first time that leveraging our ITL training
scheme is able to alleviate challenging catastrophic forgetting problems in
incremental learning. We conduct experiments using five challenging benchmark
datasets to validate the effectiveness of our incremental-transfer learning
approach. Our approach makes minimal assumptions on computation resources and
domain-specific expertise, and hence constitutes a strong starting point in
multi-site medical image segmentation.
|
110,856 | 110,856 | Slot Order Matters for Compositional Scene Understanding | Empowering agents with a compositional understanding of their environment is
a promising next step toward solving long-horizon planning problems. On the one
hand, we have seen encouraging progress on variational inference algorithms for
obtaining sets of object-centric latent representations ("slots") from
unstructured scene observations. On the other hand, generating scenes from
slots has received less attention, in part because it is complicated by the
lack of a canonical object order. A canonical object order is useful for
learning the object correlations necessary to generate physically plausible
scenes similar to how raster scan order facilitates learning pixel correlations
for pixel-level autoregressive image generation. In this work, we address this
lack by learning a fixed object order for a hierarchical variational
autoencoder with a single level of autoregressive slots and a global scene
prior. We cast autoregressive slot inference as a set-to-sequence modeling
problem. We introduce an auxiliary loss to train the slot prior to generate
objects in a fixed order. During inference, we align a set of inferred slots to
the object order obtained from a slot prior rollout. To ensure the rolled out
objects are meaningful for the given scene, we condition the prior on an
inferred global summary of the input. Experiments on compositional environments
and ablations demonstrate that our model with global prior, inference with
aligned slot order, and auxiliary loss achieves state-of-the-art sample
quality.
|
110,857 | 110,857 | Completion Time Minimization of Fog-RAN-Assisted Federated Learning With
Rate-Splitting Transmission | This work studies federated learning (FL) over a fog radio access network, in
which multiple internet-of-things (IoT) devices cooperatively learn a shared
machine learning model by communicating with a cloud server (CS) through
distributed access points (APs). Under the assumption that the fronthaul links
connecting APs to CS have finite capacity, a rate-splitting transmission at IoT
devices (IDs) is proposed which enables hybrid edge and cloud decoding of split
uplink messages. The problem of completion time minimization for FL is tackled
by optimizing the rate-splitting transmission and fronthaul quantization
strategies along with training hyperparameters such as precision and iteration
numbers. Numerical results show that the proposed rate-splitting transmission
achieves notable gains over benchmark schemes which rely solely on edge or
cloud decoding.
|
110,858 | 110,858 | Regularization-wise double descent: Why it occurs and how to eliminate
it | The risk of overparameterized models, in particular deep neural networks, is
often double-descent shaped as a function of the model size. Recently, it was
shown that the risk as a function of the early-stopping time can also be
double-descent shaped, and this behavior can be explained as a super-position
of bias-variance tradeoffs. In this paper, we show that the risk of explicit
L2-regularized models can exhibit double descent behavior as a function of the
regularization strength, both in theory and practice. We find that for linear
regression, a double descent shaped risk is caused by a superposition of
bias-variance tradeoffs corresponding to different parts of the model and can
be mitigated by scaling the regularization strength of each part appropriately.
Motivated by this result, we study a two-layer neural network and show that
double descent can be eliminated by adjusting the regularization strengths for
the first and second layer. Lastly, we study a 5-layer CNN and ResNet-18
trained on CIFAR-10 with label noise, and CIFAR-100 without label noise, and
demonstrate that all exhibit double descent behavior as a function of the
regularization strength.
|
110,859 | 110,859 | Instant Graph Neural Networks for Dynamic Graphs | Graph Neural Networks (GNNs) have been widely used for modeling
graph-structured data. With the development of numerous GNN variants, recent
years have witnessed groundbreaking results in improving the scalability of
GNNs to work on static graphs with millions of nodes. However, how to instantly
represent continuous changes of large-scale dynamic graphs with GNNs is still
an open problem. Existing dynamic GNNs focus on modeling the periodic evolution
of graphs, often on a snapshot basis. Such methods suffer from two drawbacks:
first, there is a substantial delay for the changes in the graph to be
reflected in the graph representations, resulting in losses on the model's
accuracy; second, repeatedly calculating the representation matrix on the
entire graph in each snapshot is predominantly time-consuming and severely
limits the scalability. In this paper, we propose Instant Graph Neural Network
(InstantGNN), an incremental computation approach for the graph representation
matrix of dynamic graphs. Set to work with dynamic graphs with the edge-arrival
model, our method avoids time-consuming, repetitive computations and allows
instant updates on the representation and instant predictions. Graphs with
dynamic structures and dynamic attributes are both supported. The upper bounds
of time complexity of those updates are also provided. Furthermore, our method
provides an adaptive training strategy, which guides the model to retrain at
moments when it can make the greatest performance gains. We conduct extensive
experiments on several real-world and synthetic datasets. Empirical results
demonstrate that our model achieves state-of-the-art accuracy while having
orders-of-magnitude higher efficiency than existing methods.
|
110,860 | 110,860 | Generalization for multiclass classification with overparameterized
linear models | Via an overparameterized linear model with Gaussian features, we provide
conditions for good generalization for multiclass classification of
minimum-norm interpolating solutions in an asymptotic setting where both the
number of underlying features and the number of classes scale with the number
of training points. The survival/contamination analysis framework for
understanding the behavior of overparameterized learning problems is adapted to
this setting, revealing that multiclass classification qualitatively behaves
like binary classification in that, as long as there are not too many classes
(made precise in the paper), it is possible to generalize well even in some
settings where the corresponding regression tasks would not generalize. Besides
various technical challenges, it turns out that the key difference from the
binary classification setting is that there are relatively fewer positive
training examples of each class in the multiclass setting as the number of
classes increases, making the multiclass problem "harder" than the binary one.
|
110,861 | 110,861 | MetaLR: Layer-wise Learning Rate based on Meta-Learning for Adaptively
Fine-tuning Medical Pre-trained Models | When applying transfer learning for medical image analysis, downstream tasks
often have significant gaps with the pre-training tasks. Previous methods
mainly focus on improving the transferabilities of the pre-trained models to
bridge the gaps. In fact, model fine-tuning can also play a very important role
in tackling this problem. A conventional fine-tuning method is updating all
deep neural networks (DNNs) layers by a single learning rate (LR), which
ignores the unique transferabilities of different layers. In this work, we
explore the behaviors of different layers in the fine-tuning stage. More
precisely, we first hypothesize that lower-level layers are more
domain-specific while higher-level layers are more task-specific, which is
verified by a simple bi-directional fine-tuning scheme. It is harder for the
pre-trained specific layers to transfer to new tasks than general layers. On
this basis, to make different layers better co-adapt to the downstream tasks
according to their transferabilities, a meta-learning-based LR learner, namely
MetaLR, is proposed to assign LRs for each layer automatically. Extensive
experiments on various medical applications (i.e., POCUS, BUSI, Chest X-ray,
and LiTS) well confirm our hypothesis and show the superior performance of the
proposed methods to previous state-of-the-art fine-tuning methods.
|
110,862 | 110,862 | Hybrid Models for Mixed Variables in Bayesian Optimization | We systematically describe the problem of simultaneous surrogate modeling of
mixed variables (i.e., continuous, integer and categorical variables) in the
Bayesian optimization (BO) context. We provide a unified hybrid model using
both Monte-Carlo tree search (MCTS) and Gaussian processes (GP) that
encompasses and generalizes multiple state-of-the-art mixed BO surrogates.
Based on the architecture, we propose applying a new dynamic model selection
criterion among novel candidate families of covariance kernels, including
non-stationary kernels and associated families. Different benchmark problems
are studied and presented to support the superiority of our model, along with
results highlighting the effectiveness of our method compared to most
state-of-the-art mixed-variable methods in BO.
|
110,863 | 110,863 | Fair Classification via Transformer Neural Networks: Case Study of an
Educational Domain | Educational technologies nowadays increasingly use data and Machine Learning
(ML) models. This gives the students, instructors, and administrators support
and insights for the optimum policy. However, it is well acknowledged that ML
models are subject to bias, which raises concern about the fairness, bias, and
discrimination of using these automated ML algorithms in education and its
unintended and unforeseen negative consequences. The contribution of bias
during the decision-making comes from datasets used for training ML models and
the model architecture. This paper presents a preliminary investigation of
fairness constraint in transformer neural networks on Law School and
Student-Mathematics datasets. The used transformer models transform these raw
datasets into a richer representation space of natural language processing
(NLP) while solving fairness classification. We have employed fairness metrics
for evaluation and check the trade-off between fairness and accuracy. We have
reported the various metrics of F1, SPD, EOD, and accuracy for different
architectures from the transformer model class.
|
110,864 | 110,864 | Impact of the composition of feature extraction and class sampling in
medicare fraud detection | With healthcare being critical aspect, health insurance has become an
important scheme in minimizing medical expenses. Following this, the healthcare
industry has seen a significant increase in fraudulent activities owing to
increased insurance, and fraud has become a significant contributor to rising
medical care expenses, although its impact can be mitigated using fraud
detection techniques. To detect fraud, machine learning techniques are used.
The Centers for Medicaid and Medicare Services (CMS) of the United States
federal government released "Medicare Part D" insurance claims is utilized in
this study to develop fraud detection system. Employing machine learning
algorithms on a class-imbalanced and high dimensional medicare dataset is a
challenging task. To compact such challenges, the present work aims to perform
feature extraction following data sampling, afterward applying various
classification algorithms, to get better performance. Feature extraction is a
dimensionality reduction approach that converts attributes into linear or
non-linear combinations of the actual attributes, generating a smaller and more
diversified set of attributes and thus reducing the dimensions. Data sampling
is commonlya used to address the class imbalance either by expanding the
frequency of minority class or reducing the frequency of majority class to
obtain approximately equal numbers of occurrences for both classes. The
proposed approach is evaluated through standard performance metrics. Thus, to
detect fraud efficiently, this study applies autoencoder as a feature
extraction technique, synthetic minority oversampling technique (SMOTE) as a
data sampling technique, and various gradient boosted decision tree-based
classifiers as a classification algorithm. The experimental results show the
combination of autoencoders followed by SMOTE on the LightGBM classifier
achieved best results.
|
110,865 | 110,865 | Rate-Optimal Online Convex Optimization in Adaptive Linear Control | We consider the problem of controlling an unknown linear dynamical system
under adversarially changing convex costs and full feedback of both the state
and cost function. We present the first computationally-efficient algorithm
that attains an optimal $\smash{\sqrt{T}}$-regret rate compared to the best
stabilizing linear controller in hindsight, while avoiding stringent
assumptions on the costs such as strong convexity. Our approach is based on a
careful design of non-convex lower confidence bounds for the online costs, and
uses a novel technique for computationally-efficient regret minimization of
these bounds that leverages their particular non-convex structure.
|
110,866 | 110,866 | On the Generalization of Wasserstein Robust Federated Learning | In federated learning, participating clients typically possess non-i.i.d.
data, posing a significant challenge to generalization to unseen distributions.
To address this, we propose a Wasserstein distributionally robust optimization
scheme called WAFL. Leveraging its duality, we frame WAFL as an empirical
surrogate risk minimization problem, and solve it using a local SGD-based
algorithm with convergence guarantees. We show that the robustness of WAFL is
more general than related approaches, and the generalization bound is robust to
all adversarial distributions inside the Wasserstein ball (ambiguity set).
Since the center location and radius of the Wasserstein ball can be suitably
modified, WAFL shows its applicability not only in robustness but also in
domain adaptation. Through empirical evaluation, we demonstrate that WAFL
generalizes better than the vanilla FedAvg in non-i.i.d. settings, and is more
robust than other related methods in distribution shift settings. Further,
using benchmark datasets we show that WAFL is capable of generalizing to unseen
target domains.
|
110,867 | 110,867 | Modeling electronic health record data using a knowledge-graph-embedded
topic model | The rapid growth of electronic health record (EHR) datasets opens up
promising opportunities to understand human diseases in a systematic way.
However, effective extraction of clinical knowledge from the EHR data has been
hindered by its sparsity and noisy information. We present KG-ETM, an
end-to-end knowledge graph-based multimodal embedded topic model. KG-ETM
distills latent disease topics from EHR data by learning the embedding from the
medical knowledge graphs. We applied KG-ETM to a large-scale EHR dataset
consisting of over 1 million patients. We evaluated its performance based on
EHR reconstruction and drug imputation. KG-ETM demonstrated superior
performance over the alternative methods on both tasks. Moreover, our model
learned clinically meaningful graph-informed embedding of the EHR codes. In
additional, our model is also able to discover interpretable and accurate
patient representations for patient stratification and drug recommendations.
|
110,868 | 110,868 | XPASC: Measuring Generalization in Weak Supervision | Weak supervision is leveraged in a wide range of domains and tasks due to its
ability to create massive amounts of labeled data, requiring only little manual
effort. Standard approaches use labeling functions to specify signals that are
relevant for the labeling. It has been conjectured that weakly supervised
models over-rely on those signals and as a result suffer from overfitting. To
verify this assumption, we introduce a novel method, XPASC
(eXPlainability-Association SCore), for measuring the generalization of a model
trained with a weakly supervised dataset. Considering the occurrences of
features, classes and labeling functions in a dataset, XPASC takes into account
the relevance of each feature for the predictions of the model as well as the
associations of the feature with the class and the labeling function,
respectively. The association in XPASC can be measured in two variants:
XPASC-CHI SQAURE measures associations relative to their statistical
significance, while XPASC-PPMI measures association strength more generally.
We use XPASC to analyze KnowMAN, an adversarial architecture intended to
control the degree of generalization from the labeling functions and thus to
mitigate the problem of overfitting. On one hand, we show that KnowMAN is able
to control the degree of generalization through a hyperparameter. On the other
hand, results and qualitative analysis show that generalization and performance
do not relate one-to-one, and that the highest degree of generalization does
not necessarily imply the best performance. Therefore methods that allow for
controlling the amount of generalization can achieve the right degree of benign
overfitting. Our contributions in this study are i) the XPASC score to measure
generalization in weakly-supervised models, ii) evaluation of XPASC across
datasets and models and iii) the release of the XPASC implementation.
|
110,869 | 110,869 | Indirect Active Learning | Traditional models of active learning assume a learner can directly
manipulate or query a covariate $X$ in order to study its relationship with a
response $Y$. However, if $X$ is a feature of a complex system, it may be
possible only to indirectly influence $X$ by manipulating a control variable
$Z$, a scenario we refer to as Indirect Active Learning. Under a nonparametric
model of Indirect Active Learning with a fixed budget, we study minimax
convergence rates for estimating the relationship between $X$ and $Y$ locally
at a point, obtaining different rates depending on the complexities and noise
levels of the relationships between $Z$ and $X$ and between $X$ and $Y$. We
also identify minimax rates for passive learning under comparable assumptions.
In many cases, our results show that, while there is an asymptotic benefit to
active learning, this benefit is fully realized by a simple two-stage learner
that runs two passive experiments in sequence.
|
110,870 | 110,870 | Safety Certification for Stochastic Systems via Neural Barrier Functions | Providing non-trivial certificates of safety for non-linear stochastic
systems is an important open problem that limits the wider adoption of
autonomous systems in safety-critical applications. One promising solution to
address this problem is barrier functions. The composition of a barrier
function with a stochastic system forms a supermartingale, thus enabling the
computation of the probability that the system stays in a safe set over a
finite time horizon via martingale inequalities. However, existing approaches
to find barrier functions for stochastic systems generally rely on convex
optimization programs that restrict the search of a barrier to a small class of
functions such as low degree SoS polynomials and can be computationally
expensive. In this paper, we parameterize a barrier function as a neural
network and show that techniques for robust training of neural networks can be
successfully employed to find neural barrier functions. Specifically, we
leverage bound propagation techniques to certify that a neural network
satisfies the conditions to be a barrier function via linear programming and
then employ the resulting bounds at training time to enforce the satisfaction
of these conditions. We also present a branch-and-bound scheme that makes the
certification framework scalable. We show that our approach outperforms
existing methods in several case studies and often returns certificates of
safety that are orders of magnitude larger.
|
110,871 | 110,871 | PAC Statistical Model Checking of Mean Payoff in Discrete- and
Continuous-Time MDP | Markov decision processes (MDP) and continuous-time MDP (CTMDP) are the
fundamental models for non-deterministic systems with probabilistic
uncertainty. Mean payoff (a.k.a. long-run average reward) is one of the most
classic objectives considered in their context. We provide the first algorithm
to compute mean payoff probably approximately correctly in unknown MDP;
further, we extend it to unknown CTMDP. We do not require any knowledge of the
state space, only a lower bound on the minimum transition probability, which
has been advocated in literature. In addition to providing probably
approximately correct (PAC) bounds for our algorithm, we also demonstrate its
practical nature by running experiments on standard benchmarks.
|
110,872 | 110,872 | Zero-Shot Bird Species Recognition by Learning from Field Guides | We exploit field guides to learn bird species recognition, in particular
zero-shot recognition of unseen species. The illustrations contained in field
guides deliberately focus on discriminative properties of a species, and can
serve as side information to transfer knowledge from seen to unseen classes. We
study two approaches: (1) a contrastive encoding of illustrations that can be
fed into zero-shot learning schemes; and (2) a novel method that leverages the
fact that illustrations are also images and as such structurally more similar
to photographs than other kinds of side information. Our results show that
illustrations from field guides, which are readily available for a wide range
of species, are indeed a competitive source of side information. On the
iNaturalist2021 subset, we obtain a harmonic mean from 749 seen and 739 unseen
classes greater than $45\%$ (@top-10) and $15\%$ (@top-1). Which shows that
field guides are a valuable option for challenging real-world scenarios with
many species.
|
110,873 | 110,873 | Evaluating Transfer-based Targeted Adversarial Perturbations against
Real-World Computer Vision Systems based on Human Judgments | Computer vision systems are remarkably vulnerable to adversarial
perturbations. Transfer-based adversarial images are generated on one (source)
system and used to attack another (target) system. In this paper, we take the
first step to investigate transfer-based targeted adversarial images in a
realistic scenario where the target system is trained on some private data with
its inventory of semantic labels not publicly available. Our main contributions
include an extensive human-judgment-based evaluation of attack success on the
Google Cloud Vision API and additional analysis of the different behaviors of
Google Cloud Vision in face of original images vs. adversarial images.
Resources are publicly available at
\url{https://github.com/ZhengyuZhao/Targeted-Tansfer/blob/main/google_results.zip}.
|
110,874 | 110,874 | Offline Reinforcement Learning with Causal Structured World Models | Model-based methods have recently shown promising for offline reinforcement
learning (RL), aiming to learn good policies from historical data without
interacting with the environment. Previous model-based offline RL methods learn
fully connected nets as world-models that map the states and actions to the
next-step states. However, it is sensible that a world-model should adhere to
the underlying causal effect such that it will support learning an effective
policy generalizing well in unseen states. In this paper, We first provide
theoretical results that causal world-models can outperform plain world-models
for offline RL by incorporating the causal structure into the generalization
error bound. We then propose a practical algorithm, oFfline mOdel-based
reinforcement learning with CaUsal Structure (FOCUS), to illustrate the
feasibility of learning and leveraging causal structure in offline RL.
Experimental results on two benchmarks show that FOCUS reconstructs the
underlying causal structure accurately and robustly. Consequently, it performs
better than the plain model-based offline RL algorithms and other causal
model-based RL algorithms.
|
110,875 | 110,875 | Functional Connectivity Methods for EEG-based Biometrics on a Large,
Heterogeneous Dataset | This study examines the utility of functional connectivity (FC) and
graph-based (GB) measures with a support vector machine classifier for use in
electroencephalogram (EEG) based biometrics. Although FC-based features have
been used in biometric applications, studies assessing the identification
algorithms on heterogeneous and large datasets are scarce. This work
investigates the performance of FC and GB metrics on a dataset of 184 subjects
formed by pooling three datasets recorded under different protocols and
acquisition systems. The results demonstrate the higher discriminatory power of
FC than GB metrics. The identification accuracy increases with higher frequency
EEG bands, indicating the enhanced uniqueness of the neural signatures in beta
and gamma bands. Using all the 56 EEG channels common to the three databases,
the best identification accuracy of 97.4% is obtained using phase-locking value
(PLV) based measures extracted from the gamma frequency band. Further, we
investigate the effect of the length of the analysis epoch to determine the
data acquisition time required to obtain satisfactory identification accuracy.
When the number of channels is reduced to 21 from 56, there is a marginal
reduction of 2.4% only in the identification accuracy using PLV features in the
gamma band. Additional experiments have been conducted to study the effect of
the cognitive state of the subject and mismatched train/test conditions on the
performance of the system.
|
110,876 | 110,876 | Finding Rule-Interpretable Non-Negative Data Representation | Non-negative Matrix Factorization (NMF) is an intensively used technique for
obtaining parts-based, lower dimensional and non-negative representation of
non-negative data. It is a popular method in different research fields.
Scientists performing research in the fields of biology, medicine and pharmacy
often prefer NMF over other dimensionality reduction approaches (such as PCA)
because the non-negativity of the approach naturally fits the characteristics
of the domain problem and its result is easier to analyze and understand.
Despite these advantages, it still can be hard to get exact characterization
and interpretation of the NMF's resulting latent factors due to their numerical
nature. On the other hand, rule-based approaches are often considered more
interpretable but lack the parts-based interpretation. In this work, we present
a version of the NMF approach that merges rule-based descriptions with
advantages of part-based representation offered by the NMF approach. Given the
numerical input data with non-negative entries and a set of rules with high
entity coverage, the approach creates the lower-dimensional non-negative
representation of the input data in such a way that its factors are described
by the appropriate subset of the input rules. In addition to revealing
important attributes for latent factors, it allows analyzing relations between
these attributes and provides the exact numerical intervals or categorical
values they take. The proposed approach provides numerous advantages in tasks
such as focused embedding or performing supervised multi-label NMF.
|
110,877 | 110,877 | Transferring Studies Across Embodiments: A Case Study in Confusion
Detection | Human-robot studies are expensive to conduct and difficult to control, and as
such researchers sometimes turn to human-avatar interaction in the hope of
faster and cheaper data collection that can be transferred to the robot domain.
In terms of our work, we are particularly interested in the challenge of
detecting and modelling user confusion in interaction, and as part of this
research programme, we conducted situated dialogue studies to investigate
users' reactions in confusing scenarios that we give in both physical and
virtual environments. In this paper, we present a combined review of these
studies and the results that we observed across these two embodiments. For the
physical embodiment, we used a Pepper Robot, while for the virtual modality, we
used a 3D avatar. Our study shows that despite attitudinal differences and
technical control limitations, there were a number of similarities detected in
user behaviour and self-reporting results across embodiment options. This work
suggests that, while avatar interaction is no true substitute for robot
interaction studies, sufficient care in study design may allow well executed
human-avatar studies to supplement more challenging human-robot studies.
|
110,878 | 110,878 | Constraining Gaussian processes for physics-informed acoustic emission
mapping | The automated localisation of damage in structures is a challenging but
critical ingredient in the path towards predictive or condition-based
maintenance of high value structures. The use of acoustic emission time of
arrival mapping is a promising approach to this challenge, but is severely
hindered by the need to collect a dense set of artificial acoustic emission
measurements across the structure, resulting in a lengthy and often impractical
data acquisition process. In this paper, we consider the use of
physics-informed Gaussian processes for learning these maps to alleviate this
problem. In the approach, the Gaussian process is constrained to the physical
domain such that information relating to the geometry and boundary conditions
of the structure are embedded directly into the learning process, returning a
model that guarantees that any predictions made satisfy physically-consistent
behaviour at the boundary. A number of scenarios that arise when training
measurement acquisition is limited, including where training data are sparse,
and also of limited coverage over the structure of interest. Using a complex
plate-like structure as an experimental case study, we show that our approach
significantly reduces the burden of data collection, where it is seen that
incorporation of boundary condition knowledge significantly improves predictive
accuracy as training observations are reduced, particularly when training
measurements are not available across all parts of the structure.
|
110,879 | 110,879 | Causality Learning With Wasserstein Generative Adversarial Networks | Conventional methods for causal structure learning from data face significant
challenges due to combinatorial search space. Recently, the problem has been
formulated into a continuous optimization framework with an acyclicity
constraint to learn Directed Acyclic Graphs (DAGs). Such a framework allows the
utilization of deep generative models for causal structure learning to better
capture the relations between data sample distributions and DAGs. However, so
far no study has experimented with the use of Wasserstein distance in the
context of causal structure learning. Our model named DAG-WGAN combines the
Wasserstein-based adversarial loss with an acyclicity constraint in an
auto-encoder architecture. It simultaneously learns causal structures while
improving its data generation capability. We compare the performance of
DAG-WGAN with other models that do not involve the Wasserstein metric in order
to identify its contribution to causal structure learning. Our model performs
better with high cardinality data according to our experiments.
|
110,880 | 110,880 | Can Hybrid Geometric Scattering Networks Help Solve the Maximal Clique
Problem? | We propose a geometric scattering-based graph neural network (GNN) for
approximating solutions of the NP-hard maximal clique (MC) problem. We
construct a loss function with two terms, one which encourages the network to
find a large set of nodes and the other which acts as a surrogate for the
constraint that the nodes form a clique. We then use this loss to train a novel
GNN architecture that outputs a vector representing the probability for each
node to be part of the MC and apply a rule-based decoder to make our final
prediction. The incorporation of the scattering transform alleviates the
so-called oversmoothing problem that is often encountered in GNNs and would
degrade the performance of our proposed setup. Our empirical results
demonstrate that our method outperforms representative GNN baselines in terms
of solution accuracy and inference speed as well as conventional solvers like
GUROBI with limited time budgets.
|
110,881 | 110,881 | Can Requirements Engineering Support Explainable Artificial
Intelligence? Towards a User-Centric Approach for Explainability Requirements | With the recent proliferation of artificial intelligence systems, there has
been a surge in the demand for explainability of these systems. Explanations
help to reduce system opacity, support transparency, and increase stakeholder
trust. In this position paper, we discuss synergies between requirements
engineering (RE) and Explainable AI (XAI). We highlight challenges in the field
of XAI, and propose a framework and research directions on how RE practices can
help to mitigate these challenges.
|
110,882 | 110,882 | Canonical convolutional neural networks | We introduce canonical weight normalization for convolutional neural
networks. Inspired by the canonical tensor decomposition, we express the weight
tensors in so-called canonical networks as scaled sums of outer vector
products. In particular, we train network weights in the decomposed form, where
scale weights are optimized separately for each mode. Additionally, similarly
to weight normalization, we include a global scaling parameter. We study the
initialization of the canonical form by running the power method and by drawing
randomly from Gaussian or uniform distributions. Our results indicate that we
can replace the power method with cheaper initializations drawn from standard
distributions. The canonical re-parametrization leads to competitive
normalization performance on the MNIST, CIFAR10, and SVHN data sets. Moreover,
the formulation simplifies network compression. Once training has converged,
the canonical form allows convenient model-compression by truncating the
parameter sums.
|
110,883 | 110,883 | Latent Topology Induction for Understanding Contextualized
Representations | In this work, we study the representation space of contextualized embeddings
and gain insight into the hidden topology of large language models. We show
there exists a network of latent states that summarize linguistic properties of
contextualized representations. Instead of seeking alignments to existing
well-defined annotations, we infer this latent network in a fully unsupervised
way using a structured variational autoencoder. The induced states not only
serve as anchors that mark the topology (neighbors and connectivity) of the
representation manifold but also reveal the internal mechanism of encoding
sentences. With the induced network, we: (1). decompose the representation
space into a spectrum of latent states which encode fine-grained word meanings
with lexical, morphological, syntactic and semantic information; (2). show
state-state transitions encode rich phrase constructions and serve as the
backbones of the latent space. Putting the two together, we show that sentences
are represented as a traversal over the latent network where state-state
transition chains encode syntactic templates and state-word emissions fill in
the content. We demonstrate these insights with extensive experiments and
visualizations.
|
110,884 | 110,884 | Understanding deep learning via decision boundary | This paper discovers that the neural network with lower decision boundary
(DB) variability has better generalizability. Two new notions, algorithm DB
variability and $(\epsilon, \eta)$-data DB variability, are proposed to measure
the decision boundary variability from the algorithm and data perspectives.
Extensive experiments show significant negative correlations between the
decision boundary variability and the generalizability. From the theoretical
view, two lower bounds based on algorithm DB variability are proposed and do
not explicitly depend on the sample size. We also prove an upper bound of order
$\mathcal{O}\left(\frac{1}{\sqrt{m}}+\epsilon+\eta\log\frac{1}{\eta}\right)$
based on data DB variability. The bound is convenient to estimate without the
requirement of labels, and does not explicitly depend on the network size which
is usually prohibitively large in deep learning.
|
110,885 | 110,885 | A Survey on Surrogate-assisted Efficient Neural Architecture Search | Neural architecture search (NAS) has become increasingly popular in the deep
learning community recently, mainly because it can provide an opportunity to
allow interested users without rich expertise to benefit from the success of
deep neural networks (DNNs). However, NAS is still laborious and time-consuming
because a large number of performance estimations are required during the
search process of NAS, and training DNNs is computationally intensive. To solve
the major limitation of NAS, improving the efficiency of NAS is essential in
the design of NAS. This paper begins with a brief introduction to the general
framework of NAS. Then, the methods for evaluating network candidates under the
proxy metrics are systematically discussed. This is followed by a description
of surrogate-assisted NAS, which is divided into three different categories,
namely Bayesian optimization for NAS, surrogate-assisted evolutionary
algorithms for NAS, and MOP for NAS. Finally, remaining challenges and open
research questions are discussed, and promising research topics are suggested
in this emerging field.
|
110,886 | 110,886 | A High-Performance Customer Churn Prediction System based on
Self-Attention | Customer churn prediction is a challenging domain of research that
contributes to customer retention strategy. The predictive performance of
existing machine learning models, which are often adopted by churn communities,
appear to be at a bottleneck, partly due to models' poor feature extraction
capability. Therefore, a novel algorithm, a hybrid neural network with
self-attention enhancement (HNNSAE), is proposed in this paper to improve the
efficiency of feature screening and feature extraction, consequently improving
the model's predictive performance. This model consists of three main blocks.
The first block is the entity embedding layer, which is employed to process the
categorical variables transformed into 0-1 code. The second block is the
feature extractor, which extracts the significant features through the
multi-head self-attention mechanism. In addition, to improve the feature
extraction effect, we stack the residual connection neural network on
multi-head self-attention modules. The third block is a classifier, which is a
three-layer multilayer perceptron. This work conducts experiments on publicly
available dataset related to commercial bank customers. The result demonstrates
that HNNSAE significantly outperforms the other Individual Machine Learning
(IML), Ensemble Machine Learning (EML), and Deep Learning (DL) methods tested
in this paper. Furthermore, we compare the performance of the feature extractor
proposed in this paper with that of other three feature extractors and find
that the method proposed in this paper significantly outperforms other methods.
In addition, four hypotheses about model prediction performance and overfitting
risk are tested on the publicly available dataset.
|
110,887 | 110,887 | Rethinking and Scaling Up Graph Contrastive Learning: An Extremely
Efficient Approach with Group Discrimination | Graph contrastive learning (GCL) alleviates the heavy reliance on label
information for graph representation learning (GRL) via self-supervised
learning schemes. The core idea is to learn by maximising mutual information
for similar instances, which requires similarity computation between two node
instances. However, this operation can be computationally expensive. For
example, the time complexity of two commonly adopted contrastive loss functions
(i.e., InfoNCE and JSD estimator) for a node is $O(ND)$ and $O(D)$,
respectively, where $N$ is the number of nodes, and $D$ is the embedding
dimension. Additionally, GCL normally requires a large number of training
epochs to be well-trained on large-scale datasets. Inspired by an observation
of a technical defect (i.e., inappropriate usage of Sigmoid function) commonly
used in two representative GCL works, DGI and MVGRL, we revisit GCL and
introduce a new learning paradigm for self-supervised GRL, namely, Group
Discrimination (GD), and propose a novel GD-based method called Graph Group
Discrimination (GGD). Instead of similarity computation, GGD directly
discriminates two groups of summarised node instances with a simple binary
cross-entropy loss. As such, GGD only requires $O(1)$ for loss computation of a
node. In addition, GGD requires much fewer training epochs to obtain
competitive performance compared with GCL methods on large-scale datasets.
These two advantages endow GGD with the very efficient property. Extensive
experiments show that GGD outperforms state-of-the-art self-supervised methods
on 8 datasets. In particular, GGD can be trained in 0.18 seconds (6.44 seconds
including data preprocessing) on ogbn-arxiv, which is orders of magnitude
(10,000+ faster than GCL baselines} while consuming much less memory. Trained
with 9 hours on ogbn-papers100M with billion edges, GGD outperforms its GCL
counterparts in both accuracy and efficiency.
|
110,888 | 110,888 | Accelerating hydrodynamic simulations of urban drainage systems with
physics-guided machine learning | We propose and demonstrate a new approach for fast and accurate surrogate
modelling of urban drainage system hydraulics based on physics-guided machine
learning. The surrogates are trained against a limited set of simulation
results from a hydrodynamic (HiFi) model. Our approach reduces simulation times
by one to two orders of magnitude compared to a HiFi model. It is thus slower
than e.g. conceptual hydrological models, but it enables simulations of water
levels, flows and surcharges in all nodes and links of a drainage network and
thus largely preserves the level of detail provided by HiFi models. Comparing
time series simulated by the surrogate and the HiFi model, R2 values in the
order of 0.9 are achieved. Surrogate training times are currently in the order
of one hour. However, they can likely be reduced through the application of
transfer learning and graph neural networks. Our surrogate approach will be
useful for interactive workshops in initial design phases of urban drainage
systems, as well as for real time applications. In addition, our model
formulation is generic and future research should investigate its application
for simulating other water systems.
|
110,889 | 110,889 | Detecting the Severity of Major Depressive Disorder from Speech: A Novel
HARD-Training Methodology | Major Depressive Disorder (MDD) is a common worldwide mental health issue
with high associated socioeconomic costs. The prediction and automatic
detection of MDD can, therefore, make a huge impact on society. Speech, as a
non-invasive, easy to collect signal, is a promising marker to aid the
diagnosis and assessment of MDD. In this regard, speech samples were collected
as part of the Remote Assessment of Disease and Relapse in Major Depressive
Disorder (RADAR-MDD) research programme. RADAR-MDD was an observational cohort
study in which speech and other digital biomarkers were collected from a cohort
of individuals with a history of MDD in Spain, United Kingdom and the
Netherlands. In this paper, the RADAR-MDD speech corpus was taken as an
experimental framework to test the efficacy of a Sequence-to-Sequence model
with a local attention mechanism in a two-class depression severity
classification paradigm. Additionally, a novel training method, HARD-Training,
is proposed. It is a methodology based on the selection of more ambiguous
samples for the model training, and inspired by the curriculum learning
paradigm. HARD-Training was found to consistently improve - with an average
increment of 8.6% - the performance of our classifiers for both of two speech
elicitation tasks used and each collection site of the RADAR-MDD speech corpus.
With this novel methodology, our Sequence-to-Sequence model was able to
effectively detect MDD severity regardless of language. Finally, recognising
the need for greater awareness of potential algorithmic bias, we conduct an
additional analysis of our results separately for each gender.
|
110,890 | 110,890 | Beyond Opinion Mining: Summarizing Opinions of Customer Reviews | Customer reviews are vital for making purchasing decisions in the Information
Age. Such reviews can be automatically summarized to provide the user with an
overview of opinions. In this tutorial, we present various aspects of opinion
summarization that are useful for researchers and practitioners. First, we will
introduce the task and major challenges. Then, we will present existing opinion
summarization solutions, both pre-neural and neural. We will discuss how
summarizers can be trained in the unsupervised, few-shot, and supervised
regimes. Each regime has roots in different machine learning methods, such as
auto-encoding, controllable text generation, and variational inference.
Finally, we will discuss resources and evaluation methods and conclude with the
future directions. This three-hour tutorial will provide a comprehensive
overview over major advances in opinion summarization. The listeners will be
well-equipped with the knowledge that is both useful for research and practical
applications.
|
110,891 | 110,891 | Truly Mesh-free Physics-Informed Neural Networks | Physics-informed Neural Networks (PINNs) have recently emerged as a
principled way to include prior physical knowledge in form of partial
differential equations (PDEs) into neural networks. Although generally viewed
as being mesh-free, current approaches still rely on collocation points
obtained within a bounded region, even in settings with spatially sparse
signals. Furthermore, if the boundaries are not known, the selection of such a
region may be arbitrary, resulting in a large proportion of collocation points
being selected in areas of low relevance. To resolve this, we present a
mesh-free and adaptive approach termed particle-density PINN (pdPINN), which is
inspired by the microscopic viewpoint of fluid dynamics. Instead of sampling
from a bounded region, we propose to sample directly from the distribution over
the (fluids) particle positions, eliminating the need to introduce boundaries
while adaptively focusing on the most relevant regions. This is achieved by
reformulating the modeled fluid density as an unnormalized probability
distribution from which we sample with dynamic Monte Carlo methods. We further
generalize pdPINNs to different settings that allow interpreting a positive
scalar quantity as a particle density, such as the evolution of the temperature
in the heat equation. The utility of our approach is demonstrated on
experiments for modeling (non-steady) compressible fluids in up to three
dimensions and a two-dimensional diffusion problem, illustrating the high
flexibility and sample efficiency compared to existing refinement methods for
PINNs.
|
110,892 | 110,892 | Is an encoder within reach? | The encoder network of an autoencoder is an approximation of the nearest
point projection onto the manifold spanned by the decoder. A concern with this
approximation is that, while the output of the encoder is always unique, the
projection can possibly have infinitely many values. This implies that the
latent representations learned by the autoencoder can be misleading. Borrowing
from geometric measure theory, we introduce the idea of using the reach of the
manifold spanned by the decoder to determine if an optimal encoder exists for a
given dataset and decoder. We develop a local generalization of this reach and
propose a numerical estimator thereof. We demonstrate that this allows us to
determine which observations can be expected to have a unique, and thereby
trustworthy, latent representation. As our local reach estimator is
differentiable, we investigate its usage as a regularizer and show that this
leads to learned manifolds for which projections are more often unique than
without regularization.
|
110,893 | 110,893 | Disentangling Epistemic and Aleatoric Uncertainty in Reinforcement
Learning | Characterizing aleatoric and epistemic uncertainty on the predicted rewards
can help in building reliable reinforcement learning (RL) systems. Aleatoric
uncertainty results from the irreducible environment stochasticity leading to
inherently risky states and actions. Epistemic uncertainty results from the
limited information accumulated during learning to make informed decisions.
Characterizing aleatoric and epistemic uncertainty can be used to speed up
learning in a training environment, improve generalization to similar testing
environments, and flag unfamiliar behavior in anomalous testing environments.
In this work, we introduce a framework for disentangling aleatoric and
epistemic uncertainty in RL. (1) We first define four desiderata that capture
the desired behavior for aleatoric and epistemic uncertainty estimation in RL
at both training and testing time. (2) We then present four RL models inspired
by supervised learning (i.e. Monte Carlo dropout, ensemble, deep kernel
learning models, and evidential networks) to instantiate aleatoric and
epistemic uncertainty. Finally, (3) we propose a practical evaluation method to
evaluate uncertainty estimation in model-free RL based on detection of
out-of-distribution environments and generalization to perturbed environments.
We present theoretical and experimental evidence to validate that carefully
equipping model-free RL agents with supervised learning uncertainty methods can
fulfill our desiderata.
|
110,894 | 110,894 | Prescriptive maintenance with causal machine learning | Machine maintenance is a challenging operational problem, where the goal is
to plan sufficient preventive maintenance to avoid machine failures and
overhauls. Maintenance is often imperfect in reality and does not make the
asset as good as new. Although a variety of imperfect maintenance policies have
been proposed in the literature, these rely on strong assumptions regarding the
effect of maintenance on the machine's condition, assuming the effect is (1)
deterministic or governed by a known probability distribution, and (2)
machine-independent. This work proposes to relax both assumptions by learning
the effect of maintenance conditional on a machine's characteristics from
observational data on similar machines using existing methodologies for causal
inference. By predicting the maintenance effect, we can estimate the number of
overhauls and failures for different levels of maintenance and, consequently,
optimize the preventive maintenance frequency to minimize the total estimated
cost. We validate our proposed approach using real-life data on more than 4,000
maintenance contracts from an industrial partner. Empirical results show that
our novel, causal approach accurately predicts the maintenance effect and
results in individualized maintenance schedules that are more accurate and
cost-effective than supervised or non-individualized approaches.
|
110,895 | 110,895 | Optimal Weak to Strong Learning | The classic algorithm AdaBoost allows to convert a weak learner, that is an
algorithm that produces a hypothesis which is slightly better than chance, into
a strong learner, achieving arbitrarily high accuracy when given enough
training data. We present a new algorithm that constructs a strong learner from
a weak learner but uses less training data than AdaBoost and all other weak to
strong learners to achieve the same generalization bounds. A sample complexity
lower bound shows that our new algorithm uses the minimum possible amount of
training data and is thus optimal. Hence, this work settles the sample
complexity of the classic problem of constructing a strong learner from a weak
learner.
|
110,896 | 110,896 | On Calibration of Graph Neural Networks for Node Classification | Graphs can model real-world, complex systems by representing entities and
their interactions in terms of nodes and edges. To better exploit the graph
structure, graph neural networks have been developed, which learn entity and
edge embeddings for tasks such as node classification and link prediction.
These models achieve good performance with respect to accuracy, but the
confidence scores associated with the predictions might not be calibrated. That
means that the scores might not reflect the ground-truth probabilities of the
predicted events, which would be especially important for safety-critical
applications. Even though graph neural networks are used for a wide range of
tasks, the calibration thereof has not been sufficiently explored yet. We
investigate the calibration of graph neural networks for node classification,
study the effect of existing post-processing calibration methods, and analyze
the influence of model capacity, graph density, and a new loss function on
calibration. Further, we propose a topology-aware calibration method that takes
the neighboring nodes into account and yields improved calibration compared to
baseline methods.
|
110,897 | 110,897 | Decentralized Optimistic Hyperpolicy Mirror Descent: Provably No-Regret
Learning in Markov Games | We study decentralized policy learning in Markov games where we control a
single agent to play with nonstationary and possibly adversarial opponents. Our
goal is to develop a no-regret online learning algorithm that (i) takes actions
based on the local information observed by the agent and (ii) is able to find
the best policy in hindsight. For such a problem, the nonstationary state
transitions due to the varying opponent pose a significant challenge. In light
of a recent hardness result \citep{liu2022learning}, we focus on the setting
where the opponent's previous policies are revealed to the agent for decision
making. With such an information structure, we propose a new algorithm,
\underline{D}ecentralized \underline{O}ptimistic hype\underline{R}policy
m\underline{I}rror de\underline{S}cent (DORIS), which achieves
$\sqrt{K}$-regret in the context of general function approximation, where $K$
is the number of episodes. Moreover, when all the agents adopt DORIS, we prove
that their mixture policy constitutes an approximate coarse correlated
equilibrium. In particular, DORIS maintains a \textit{hyperpolicy} which is a
distribution over the policy space. The hyperpolicy is updated via mirror
descent, where the update direction is obtained by an optimistic variant of
least-squares policy evaluation. Furthermore, to illustrate the power of our
method, we apply DORIS to constrained and vector-valued MDPs, which can be
formulated as zero-sum Markov games with a fictitious opponent.
|
110,898 | 110,898 | MCD: Marginal Contrastive Discrimination for conditional density
estimation | We consider the problem of conditional density estimation, which is a major
topic of interest in the fields of statistical and machine learning. Our
method, called Marginal Contrastive Discrimination, MCD, reformulates the
conditional density function into two factors, the marginal density function of
the target variable and a ratio of density functions which can be estimated
through binary classification. Like noise-contrastive methods, MCD can leverage
state-of-the-art supervised learning techniques to perform conditional density
estimation, including neural networks. Our benchmark reveals that our method
significantly outperforms in practice existing methods on most density models
and regression datasets.
|
110,899 | 110,899 | Non-Intrusive Reduced Models based on Operator Inference for Chaotic
Systems | This work explores the physics-driven machine learning technique Operator
Inference (OpInf) for predicting the state of chaotic dynamical systems. OpInf
provides a non-intrusive approach to infer approximations of polynomial
operators in reduced space without having access to the full order operators
appearing in discretized models. Datasets for the physics systems are generated
using conventional numerical solvers and then projected to a low-dimensional
space via Principal Component Analysis (PCA). In latent space, a least-squares
problem is set to fit a quadratic polynomial operator which is subsequently
employed in a time-integration scheme in order to produce extrapolations in the
same space. Once solved, the inverse PCA operation is applied for
reconstructing the extrapolations in the original space. The quality of the
OpInf predictions is assessed via the Normalized Root Mean Squared Error
(NRMSE) metric from which the Valid Prediction Time (VPT) is computed.
Numerical experiments considering the chaotic systems Lorenz 96 and the
Kuramoto-Sivashinsky equation show promising forecasting capabilities of the
OpInf reduced order models with VPT ranges that outperform state-of-the-art
machine learning methods such as backpropagation and reservoir computing
recurrent neural networks [1], as well as Markov neural operators [2]. The best
results based on randomized initial conditions show that Lorenz 96 system can
be forecasted up to 6.66 or 3.19 Lyapunov time units corresponding to the
forcing terms F=8 and F=10, respectively, while the KS system achieved
remarkable 794 Lyapunov time units.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.