Unnamed: 0
int64 0
5k
| title
stringlengths 9
210
| abstract
stringlengths 164
1.92k
|
|---|---|---|
2,600 | Shuffled Check-in: Privacy Amplification towards Practical Distributed Learning | Recent studies of distributed computation with formal privacy guarantees,
such as differentially private (DP) federated learning, leverage random
sampling of clients in each round (privacy amplification by subsampling) to
achieve satisfactory levels of privacy. Achieving this however requires strong
assumptions which may not hold in practice, including precise and uniform
subsampling of clients, and a highly trusted aggregator to process clients'
data. In this paper, we explore a more practical protocol, shuffled check-in,
to resolve the aforementioned issues. The protocol relies on client making
independent and random decision to participate in the computation, freeing the
requirement of server-initiated subsampling, and enabling robust modelling of
client dropouts. Moreover, a weaker trust model known as the shuffle model is
employed instead of using a trusted aggregator. To this end, we introduce new
tools to characterize the R\'enyi differential privacy (RDP) of shuffled
check-in. We show that our new techniques improve at least three times in
privacy guarantee over those using approximate DP's strong composition at
various parameter regimes. Furthermore, we provide a numerical approach to
track the privacy of generic shuffled check-in mechanism including distributed
stochastic gradient descent (SGD) with Gaussian mechanism. To the best of our
knowledge, this is also the first evaluation of Gaussian mechanism within the
local/shuffle model under the distributed setting in the literature, which can
be of independent interest. |
2,601 | Group Meritocratic Fairness in Linear Contextual Bandits | We study the linear contextual bandit problem where an agent has to select
one candidate from a pool and each candidate belongs to a sensitive group. In
this setting, candidates' rewards may not be directly comparable between
groups, for example when the agent is an employer hiring candidates from
different ethnic groups and some groups have a lower reward due to
discriminatory bias and/or social injustice. We propose a notion of fairness
that states that the agent's policy is fair when it selects a candidate with
highest relative rank, which measures how good the reward is when compared to
candidates from the same group. This is a very strong notion of fairness, since
the relative rank is not directly observed by the agent and depends on the
underlying reward model and on the distribution of rewards. Thus we study the
problem of learning a policy which approximates a fair policy under the
condition that the contexts are independent between groups and the distribution
of rewards of each group is absolutely continuous. In particular, we design a
greedy policy which at each round constructs a ridge regression estimator from
the observed context-reward pairs, and then computes an estimate of the
relative rank of each candidate using the empirical cumulative distribution
function. We prove that the greedy policy achieves, after $T$ rounds, up to log
factors and with high probability, a fair pseudo-regret of order $\sqrt{dT}$,
where $d$ is the dimension of the context vectors. The policy also satisfies
demographic parity at each round when averaged over all possible information
available before the selection. We finally show with a proof of concept
simulation that our policy achieves sub-linear fair pseudo-regret also in
practice. |
2,602 | Transfer learning to decode brain states reflecting the relationship between cognitive tasks | Transfer learning improves the performance of the target task by leveraging
the data of a specific source task: the closer the relationship between the
source and the target tasks, the greater the performance improvement by
transfer learning. In neuroscience, the relationship between cognitive tasks is
usually represented by similarity of activated brain regions or neural
representation. However, no study has linked transfer learning and neuroscience
to reveal the relationship between cognitive tasks. In this study, we propose a
transfer learning framework to reflect the relationship between cognitive
tasks, and compare the task relations reflected by transfer learning and by the
overlaps of brain regions (e.g., neurosynth). Our results of transfer learning
create cognitive taskonomy to reflect the relationship between cognitive tasks
which is well in line with the task relations derived from neurosynth. Transfer
learning performs better in task decoding with fMRI data if the source and
target cognitive tasks activate similar brain regions. Our study uncovers the
relationship of multiple cognitive tasks and provides guidance for source task
selection in transfer learning for neural decoding based on small-sample data. |
2,603 | Intra-agent speech permits zero-shot task acquisition | Human language learners are exposed to a trickle of informative,
context-sensitive language, but a flood of raw sensory data. Through both
social language use and internal processes of rehearsal and practice, language
learners are able to build high-level, semantic representations that explain
their perceptions. Here, we take inspiration from such processes of "inner
speech" in humans (Vygotsky, 1934) to better understand the role of intra-agent
speech in embodied behavior. First, we formally pose intra-agent speech as a
semi-supervised problem and develop two algorithms that enable visually
grounded captioning with little labeled language data. We then experimentally
compute scaling curves over different amounts of labeled data and compare the
data efficiency against a supervised learning baseline. Finally, we incorporate
intra-agent speech into an embodied, mobile manipulator agent operating in a 3D
virtual world, and show that with as few as 150 additional image captions,
intra-agent speech endows the agent with the ability to manipulate and answer
questions about a new object without any related task-directed experience
(zero-shot). Taken together, our experiments suggest that modelling intra-agent
speech is effective in enabling embodied agents to learn new tasks efficiently
and without direct interaction experience. |
2,604 | CitySpec: An Intelligent Assistant System for Requirement Specification in Smart Cities | An increasing number of monitoring systems have been developed in smart
cities to ensure that real-time operations of a city satisfy safety and
performance requirements. However, many existing city requirements are written
in English with missing, inaccurate, or ambiguous information. There is a high
demand for assisting city policy makers in converting human-specified
requirements to machine-understandable formal specifications for monitoring
systems. To tackle this limitation, we build CitySpec, the first intelligent
assistant system for requirement specification in smart cities. To create
CitySpec, we first collect over 1,500 real-world city requirements across
different domains from over 100 cities and extract city-specific knowledge to
generate a dataset of city vocabulary with 3,061 words. We also build a
translation model and enhance it through requirement synthesis and develop a
novel online learning framework with validation under uncertainty. The
evaluation results on real-world city requirements show that CitySpec increases
the sentence-level accuracy of requirement specification from 59.02% to 86.64%,
and has strong adaptability to a new city and a new domain (e.g., F1 score for
requirements in Seattle increases from 77.6% to 93.75% with online learning). |
2,605 | Towards Meta-learned Algorithm Selection using Implicit Fidelity Information | Automatically selecting the best performing algorithm for a given dataset or
ranking multiple of them by their expected performance supports users in
developing new machine learning applications. Most approaches for this problem
rely on dataset meta-features and landmarking performances to capture the
salient topology of the datasets and those topologies that the algorithms
attend to. Landmarking usually exploits cheap algorithms not necessarily in the
pool of candidate algorithms to get inexpensive approximations of the topology.
While somewhat indicative, handcrafted dataset meta-features and landmarks are
likely insufficient descriptors, strongly depending on the alignment of the
geometries the landmarks and candidates search for. We propose IMFAS, a method
to exploit multi-fidelity landmarking information directly from the candidate
algorithms in the form of non-parametrically non-myopic meta-learned learning
curves via LSTM networks in a few-shot setting during testing. Using this
mechanism, IMFAS jointly learns the topology of of the datasets and the
inductive biases of algorithms without expensively training them to
convergence. IMFAS produces informative landmarks, easily enriched by arbitrary
meta-features at a low computational cost, capable of producing the desired
ranking using cheaper fidelities. We additionally show that it is able to beat
Successive Halving with at most half the fidelity sequence during test time |
2,606 | Spatial-Temporal Adaptive Graph Convolution with Attention Network for Traffic Forecasting | Traffic forecasting is one canonical example of spatial-temporal learning
task in Intelligent Traffic System. Existing approaches capture spatial
dependency with a pre-determined matrix in graph convolution neural operators.
However, the explicit graph structure losses some hidden representations of
relationships among nodes. Furthermore, traditional graph convolution neural
operators cannot aggregate long-range nodes on the graph. To overcome these
limits, we propose a novel network, Spatial-Temporal Adaptive graph convolution
with Attention Network (STAAN) for traffic forecasting. Firstly, we adopt an
adaptive dependency matrix instead of using a pre-defined matrix during GCN
processing to infer the inter-dependencies among nodes. Secondly, we integrate
PW-attention based on graph attention network which is designed for global
dependency, and GCN as spatial block. What's more, a stacked dilated 1D
convolution, with efficiency in long-term prediction, is adopted in our
temporal block for capturing the different time series. We evaluate our STAAN
on two real-world datasets, and experiments validate that our model outperforms
state-of-the-art baselines. |
2,607 | Signal Propagation in Transformers: Theoretical Perspectives and the Role of Rank Collapse | Transformers have achieved remarkable success in several domains, ranging
from natural language processing to computer vision. Nevertheless, it has been
recently shown that stacking self-attention layers - the distinctive
architectural component of Transformers - can result in rank collapse of the
tokens' representations at initialization. The question of if and how rank
collapse affects training is still largely unanswered, and its investigation is
necessary for a more comprehensive understanding of this architecture. In this
work, we shed new light on the causes and the effects of this phenomenon.
First, we show that rank collapse of the tokens' representations hinders
training by causing the gradients of the queries and keys to vanish at
initialization. Furthermore, we provide a thorough description of the origin of
rank collapse and discuss how to prevent it via an appropriate depth-dependent
scaling of the residual branches. Finally, our analysis unveils that specific
architectural hyperparameters affect the gradients of queries and values
differently, leading to disproportionate gradient norms. This suggests an
explanation for the widespread use of adaptive methods for Transformers'
optimization. |
2,608 | Singapore Soundscape Site Selection Survey (S5): Identification of Characteristic Soundscapes of Singapore via Weighted k-means Clustering | The ecological validity of soundscape studies usually rests on a choice of
soundscapes that are representative of the perceptual space under
investigation. For example, a soundscape pleasantness study might investigate
locations with soundscapes ranging from "pleasant" to "annoying". The choice of
soundscapes is typically researcher-led, but a participant-led process can
reduce selection bias and improve result reliability. Hence, we propose a
robust participant-led method to pinpoint characteristic soundscapes possessing
arbitrary perceptual attributes. We validate our method by identifying
Singaporean soundscapes spanning the perceptual quadrants generated from the
"Pleasantness" and "Eventfulness" axes of the ISO 12913-2 circumplex model of
soundscape perception, as perceived by local experts. From memory and
experience, 67 participants first selected locations corresponding to each
perceptual quadrant in each major planning region of Singapore. We then
performed weighted k-means clustering on the selected locations, with weights
for each location derived from previous frequencies and durations spent in each
location by each participant. Weights hence acted as proxies for participant
confidence. In total, 62 locations were thereby identified as suitable
locations with characteristic soundscapes for further research utilizing the
ISO 12913-2 perceptual quadrants. Audio-visual recordings and acoustic
characterization of the soundscapes will be made in a future study. |
2,609 | Better Best of Both Worlds Bounds for Bandits with Switching Costs | We study best-of-both-worlds algorithms for bandits with switching cost,
recently addressed by Rouyer, Seldin and Cesa-Bianchi, 2021. We introduce a
surprisingly simple and effective algorithm that simultaneously achieves
minimax optimal regret bound of $\mathcal{O}(T^{2/3})$ in the oblivious
adversarial setting and a bound of $\mathcal{O}(\min\{\log
(T)/\Delta^2,T^{2/3}\})$ in the stochastically-constrained regime, both with
(unit) switching costs, where $\Delta$ is the gap between the arms. In the
stochastically constrained case, our bound improves over previous results due
to Rouyer et al., that achieved regret of $\mathcal{O}(T^{1/3}/\Delta)$. We
accompany our results with a lower bound showing that, in general,
$\tilde{\Omega}(\min\{1/\Delta^2,T^{2/3}\})$ regret is unavoidable in the
stochastically-constrained case for algorithms with $\mathcal{O}(T^{2/3})$
worst-case regret. |
2,610 | Explainable Artificial Intelligence (XAI) for Internet of Things: A Survey | Black-box nature of Artificial Intelligence (AI) models do not allow users to
comprehend and sometimes trust the output created by such model. In AI
applications, where not only the results but also the decision paths to the
results are critical, such black-box AI models are not sufficient. Explainable
Artificial Intelligence (XAI) addresses this problem and defines a set of AI
models that are interpretable by the users. Recently, several number of XAI
models have been to address the issues surrounding by lack of interpretability
and explainability of black-box models in various application areas such as
healthcare, military, energy, financial and industrial domains. Although the
concept of XAI has gained great deal of attention recently, its integration
into the IoT domain has not yet been fully defined. In this paper, we provide
an in-depth and systematic review of recent studies using XAI models in the
scope of IoT domain. We categorize the studies according to their methodology
and applications areas. In addition, we aim to focus on the challenging
problems and open issues and give future directions to guide the developers and
researchers for prospective future investigations. |
2,611 | Beyond spectral gap: The role of the topology in decentralized learning | In data-parallel optimization of machine learning models, workers collaborate
to improve their estimates of the model: more accurate gradients allow them to
use larger learning rates and optimize faster. We consider the setting in which
all workers sample from the same dataset, and communicate over a sparse graph
(decentralized). In this setting, current theory fails to capture important
aspects of real-world behavior. First, the 'spectral gap' of the communication
graph is not predictive of its empirical performance in (deep) learning.
Second, current theory does not explain that collaboration enables larger
learning rates than training alone. In fact, it prescribes smaller learning
rates, which further decrease as graphs become larger, failing to explain
convergence in infinite graphs. This paper aims to paint an accurate picture of
sparsely-connected distributed optimization when workers share the same data
distribution. We quantify how the graph topology influences convergence in a
quadratic toy problem and provide theoretical results for general smooth and
(strongly) convex objectives. Our theory matches empirical observations in deep
learning, and accurately describes the relative merits of different graph
topologies. |
2,612 | A Privacy-Preserving Subgraph-Level Federated Graph Neural Network via Differential Privacy | Currently, the federated graph neural network (GNN) has attracted a lot of
attention due to its wide applications in reality without violating the privacy
regulations. Among all the privacy-preserving technologies, the differential
privacy (DP) is the most promising one due to its effectiveness and light
computational overhead. However, the DP-based federated GNN has not been well
investigated, especially in the sub-graph-level setting, such as the scenario
of recommendation system. The biggest challenge is how to guarantee the privacy
and solve the non independent and identically distributed (non-IID) data in
federated GNN simultaneously. In this paper, we propose DP-FedRec, a DP-based
federated GNN to fill the gap. Private Set Intersection (PSI) is leveraged to
extend the local graph for each client, and thus solve the non-IID problem.
Most importantly, DP is applied not only on the weights but also on the edges
of the intersection graph from PSI to fully protect the privacy of clients. The
evaluation demonstrates DP-FedRec achieves better performance with the graph
extension and DP only introduces little computations overhead. |
2,613 | An Empirical Study of IoT Security Aspects at Sentence-Level in Developer Textual Discussions | IoT is a rapidly emerging paradigm that now encompasses almost every aspect
of our modern life. As such, ensuring the security of IoT devices is crucial.
IoT devices can differ from traditional computing, thereby the design and
implementation of proper security measures can be challenging in IoT devices.
We observed that IoT developers discuss their security-related challenges in
developer forums like Stack Overflow(SO). However, we find that IoT security
discussions can also be buried inside non-security discussions in SO. In this
paper, we aim to understand the challenges IoT developers face while applying
security practices and techniques to IoT devices. We have two goals: (1)
Develop a model that can automatically find security-related IoT discussions in
SO, and (2) Study the model output to learn about IoT developer
security-related challenges. First, we download 53K posts from SO that contain
discussions about IoT. Second, we manually labeled 5,919 sentences from 53K
posts as 1 or 0. Third, we use this benchmark to investigate a suite of deep
learning transformer models. The best performing model is called SecBot.
Fourth, we apply SecBot on the entire posts and find around 30K security
related sentences. Fifth, we apply topic modeling to the security-related
sentences. Then we label and categorize the topics. Sixth, we analyze the
evolution of the topics in SO. We found that (1) SecBot is based on the
retraining of the deep learning model RoBERTa. SecBot offers the best F1-Score
of 0.935, (2) there are six error categories in misclassified samples by
SecBot. SecBot was mostly wrong when the keywords/contexts were ambiguous
(e.g., gateway can be a security gateway or a simple gateway), (3) there are 9
security topics grouped into three categories: Software, Hardware, and Network,
and (4) the highest number of topics belongs to software security, followed by
network security. |
2,614 | EiX-GNN : Concept-level eigencentrality explainer for graph neural networks | Explaining is a human knowledge transfer process regarding a phenomenon
between an explainer and an explainee. Each word used to explain this
phenomenon must be carefully selected by the explainer in accordance with the
current explainee phenomenon-related knowledge level and the phenomenon itself
in order to have a high understanding from the explainee of the phenomenon.
Nowadays, deep models, especially graph neural networks, have a major place in
daily life even in critical applications. In such context, those models need to
have a human high interpretability also referred as being explainable, in order
to improve usage trustability of them in sensitive cases. Explaining is also a
human dependent task and methods that explain deep model behavior must include
these social-related concerns for providing profitable and quality
explanations. Current explaining methods often occlude such social aspect for
providing their explanations and only focus on the signal aspect of the
question. In this contribution we propose a reliable social-aware explaining
method suited for graph neural network that includes this social feature as a
modular concept generator and by both leveraging signal and graph domain aspect
thanks to an eigencentrality concept ordering approach. Besides our method
takes into account the human-dependent aspect underlying any explanation
process, we also reach high score regarding state-of-the-art objective metrics
assessing explanation methods for graph neural networks models. |
2,615 | SubStrat: A Subset-Based Strategy for Faster AutoML | Automated machine learning (AutoML) frameworks have become important tools in
the data scientists' arsenal, as they dramatically reduce the manual work
devoted to the construction of ML pipelines. Such frameworks intelligently
search among millions of possible ML pipelines - typically containing feature
engineering, model selection and hyper parameters tuning steps - and finally
output an optimal pipeline in terms of predictive accuracy. However, when the
dataset is large, each individual configuration takes longer to execute,
therefore the overall AutoML running times become increasingly high. To this
end, we present SubStrat, an AutoML optimization strategy that tackles the data
size, rather than configuration space. It wraps existing AutoML tools, and
instead of executing them directly on the entire dataset, SubStrat uses a
genetic-based algorithm to find a small yet representative data subset which
preserves a particular characteristic of the full data. It then employs the
AutoML tool on the small subset, and finally, it refines the resulted pipeline
by executing a restricted, much shorter, AutoML process on the large dataset.
Our experimental results, performed on two popular AutoML frameworks,
Auto-Sklearn and TPOT, show that SubStrat reduces their running times by 79%
(on average), with less than 2% average loss in the accuracy of the resulted ML
pipeline. |
2,616 | Patch-based image Super Resolution using generalized Gaussian mixture model | Single Image Super Resolution (SISR) methods aim to recover the clean images
in high resolution from low resolution observations.A family of patch-based
approaches have received considerable attention and development. The minimum
mean square error (MMSE) methodis a powerful image restoration method that uses
a probability model on the patches of images. This paper proposes an algorithm
to learn a jointgeneralized Gaussian mixture model (GGMM) from a pair of the
low resolution patches and the corresponding high resolution patches fromthe
reference data. We then reconstruct the high resolution image based on the MMSE
method. Our numerical evaluations indicate that theMMSE-GGMM method competes
with other state of the art methods. |
2,617 | Recent Advances for Quantum Neural Networks in Generative Learning | Quantum computers are next-generation devices that hold promise to perform
calculations beyond the reach of classical computers. A leading method towards
achieving this goal is through quantum machine learning, especially quantum
generative learning. Due to the intrinsic probabilistic nature of quantum
mechanics, it is reasonable to postulate that quantum generative learning
models (QGLMs) may surpass their classical counterparts. As such, QGLMs are
receiving growing attention from the quantum physics and computer science
communities, where various QGLMs that can be efficiently implemented on
near-term quantum machines with potential computational advantages are
proposed. In this paper, we review the current progress of QGLMs from the
perspective of machine learning. Particularly, we interpret these QGLMs,
covering quantum circuit born machines, quantum generative adversarial
networks, quantum Boltzmann machines, and quantum autoencoders, as the quantum
extension of classical generative learning models. In this context, we explore
their intrinsic relation and their fundamental differences. We further
summarize the potential applications of QGLMs in both conventional machine
learning tasks and quantum physics. Last, we discuss the challenges and further
research directions for QGLMs. |
2,618 | Universal Speech Enhancement with Score-based Diffusion | Removing background noise from speech audio has been the subject of
considerable research and effort, especially in recent years due to the rise of
virtual communication and amateur sound recording. Yet background noise is not
the only unpleasant disturbance that can prevent intelligibility: reverb,
clipping, codec artifacts, problematic equalization, limited bandwidth, or
inconsistent loudness are equally disturbing and ubiquitous. In this work, we
propose to consider the task of speech enhancement as a holistic endeavor, and
present a universal speech enhancement system that tackles 55 different
distortions at the same time. Our approach consists of a generative model that
employs score-based diffusion, together with a multi-resolution conditioning
network that performs enhancement with mixture density networks. We show that
this approach significantly outperforms the state of the art in a subjective
test performed by expert listeners. We also show that it achieves competitive
objective scores with just 4-8 diffusion steps, despite not considering any
particular strategy for fast sampling. We hope that both our methodology and
technical contributions encourage researchers and practitioners to adopt a
universal approach to speech enhancement, possibly framing it as a generative
task. |
2,619 | Learning Backward Compatible Embeddings | Embeddings, low-dimensional vector representation of objects, are fundamental
in building modern machine learning systems. In industrial settings, there is
usually an embedding team that trains an embedding model to solve intended
tasks (e.g., product recommendation). The produced embeddings are then widely
consumed by consumer teams to solve their unintended tasks (e.g., fraud
detection). However, as the embedding model gets updated and retrained to
improve performance on the intended task, the newly-generated embeddings are no
longer compatible with the existing consumer models. This means that historical
versions of the embeddings can never be retired or all consumer teams have to
retrain their models to make them compatible with the latest version of the
embeddings, both of which are extremely costly in practice. Here we study the
problem of embedding version updates and their backward compatibility. We
formalize the problem where the goal is for the embedding team to keep updating
the embedding version, while the consumer teams do not have to retrain their
models. We develop a solution based on learning backward compatible embeddings,
which allows the embedding model version to be updated frequently, while also
allowing the latest version of the embedding to be quickly transformed into any
backward compatible historical version of it, so that consumer teams do not
have to retrain their models. Under our framework, we explore six methods and
systematically evaluate them on a real-world recommender system application. We
show that the best method, which we call BC-Aligner, maintains backward
compatibility with existing unintended tasks even after multiple model version
updates. Simultaneously, BC-Aligner achieves the intended task performance
similar to the embedding model that is solely optimized for the intended task. |
2,620 | Intelligent Circuit Design and Implementation with Machine Learning | The stagnation of EDA technologies roots from insufficient knowledge reuse.
In practice, very similar simulation or optimization results may need to be
repeatedly constructed from scratch. This motivates my research on introducing
more 'intelligence' to EDA with machine learning (ML), which explores complex
correlations in design flows based on prior data. Besides design time, I also
propose ML solutions to boost IC performance by assisting the circuit
management at runtime. In this dissertation, I present multiple fast yet
accurate ML models covering a wide range of chip design stages from the
register-transfer level (RTL) to sign-off, solving primary chip-design problems
about power, timing, interconnect, IR drop, routability, and design flow
tuning. Targeting the RTL stage, I present APOLLO, a fully automated power
modeling framework. It constructs an accurate per-cycle power model by
extracting the most power-correlated signals. The model can be further
implemented on chip for runtime power management with unprecedented low
hardware costs. Targeting gate-level netlist, I present Net2 for early
estimations on post-placement wirelength. It further enables more accurate
timing analysis without actual physical design information. Targeting circuit
layout, I present RouteNet for early routability prediction. As the first deep
learning-based routability estimator, some feature-extraction and model-design
principles proposed in it are widely adopted by later works. I also present
PowerNet for fast IR drop estimation. It captures spatial and temporal
information about power distribution with a customized CNN architecture. Last,
besides targeting a single design step, I present FIST to efficiently tune
design flow parameters during both logic synthesis and physical design. |
2,621 | How Far I'll Go: Offline Goal-Conditioned Reinforcement Learning via $f$-Advantage Regression | Offline goal-conditioned reinforcement learning (GCRL) promises
general-purpose skill learning in the form of reaching diverse goals from
purely offline datasets. We propose $\textbf{Go}$al-conditioned
$f$-$\textbf{A}$dvantage $\textbf{R}$egression (GoFAR), a novel
regression-based offline GCRL algorithm derived from a state-occupancy matching
perspective; the key intuition is that the goal-reaching task can be formulated
as a state-occupancy matching problem between a dynamics-abiding imitator agent
and an expert agent that directly teleports to the goal. In contrast to prior
approaches, GoFAR does not require any hindsight relabeling and enjoys
uninterleaved optimization for its value and policy networks. These distinct
features confer GoFAR with much better offline performance and stability as
well as statistical performance guarantee that is unattainable for prior
methods. Furthermore, we demonstrate that GoFAR's training objectives can be
re-purposed to learn an agent-independent goal-conditioned planner from purely
offline source-domain data, which enables zero-shot transfer to new target
domains. Through extensive experiments, we validate GoFAR's effectiveness in
various problem settings and tasks, significantly outperforming prior
state-of-art. Notably, on a real robotic dexterous manipulation task, while no
other method makes meaningful progress, GoFAR acquires complex manipulation
behavior that successfully accomplishes diverse goals. |
2,622 | Adaptive Weighted Nonnegative Matrix Factorization for Robust Feature Representation | Nonnegative matrix factorization (NMF) has been widely used to dimensionality
reduction in machine learning. However, the traditional NMF does not properly
handle outliers, so that it is sensitive to noise. In order to improve the
robustness of NMF, this paper proposes an adaptive weighted NMF, which
introduces weights to emphasize the different importance of each data point,
thus the algorithmic sensitivity to noisy data is decreased. It is very
different from the existing robust NMFs that use a slow growth similarity
measure. Specifically, two strategies are proposed to achieve this: fuzzier
weighted technique and entropy weighted regularized technique, and both of them
lead to an iterative solution with a simple form. Experimental results showed
that new methods have more robust feature representation on several real
datasets with noise than exsiting methods. |
2,623 | The Survival Bandit Problem | We study the survival bandit problem, a variant of the multi-armed bandit
problem introduced in an open problem by Perotto et al. (2019), with a
constraint on the cumulative reward; at each time step, the agent receives a
(possibly negative) reward and if the cumulative reward becomes lower than a
prespecified threshold, the procedure stops, and this phenomenon is called
ruin. This is the first paper studying a framework where the ruin might occur
but not always. We first discuss that a sublinear regret is unachievable under
a naive definition of the regret. Next, we provide tight lower bounds on the
probability of ruin (as well as matching policies). Based on this lower bound,
we define the survival regret as an objective to minimize and provide a policy
achieving a sublinear survival regret (at least in the case of integral
rewards) when the time horizon $T$ is known. |
2,624 | Self-Knowledge Distillation based Self-Supervised Learning for Covid-19 Detection from Chest X-Ray Images | The global outbreak of the Coronavirus 2019 (COVID-19) has overloaded
worldwide healthcare systems. Computer-aided diagnosis for COVID-19 fast
detection and patient triage is becoming critical. This paper proposes a novel
self-knowledge distillation based self-supervised learning method for COVID-19
detection from chest X-ray images. Our method can use self-knowledge of images
based on similarities of their visual features for self-supervised learning.
Experimental results show that our method achieved an HM score of 0.988, an AUC
of 0.999, and an accuracy of 0.957 on the largest open COVID-19 chest X-ray
dataset. |
2,625 | Histogram Estimation under User-level Privacy with Heterogeneous Data | We study the problem of histogram estimation under user-level differential
privacy, where the goal is to preserve the privacy of all entries of any single
user. While there is abundant literature on this classical problem under the
item-level privacy setup where each user contributes only one data point,
little has been known for the user-level counterpart. We consider the
heterogeneous scenario where both the quantity and distribution of data can be
different for each user. We propose an algorithm based on a clipping strategy
that almost achieves a two-approximation with respect to the best clipping
threshold in hindsight. This result holds without any distribution assumptions
on the data. We also prove that the clipping bias can be significantly reduced
when the counts are from non-i.i.d. Poisson distributions and show empirically
that our debiasing method provides improvements even without such constraints.
Experiments on both real and synthetic datasets verify our theoretical findings
and demonstrate the effectiveness of our algorithms. |
2,626 | Driving in Real Life with Inverse Reinforcement Learning | In this paper, we introduce the first learning-based planner to drive a car
in dense, urban traffic using Inverse Reinforcement Learning (IRL). Our
planner, DriveIRL, generates a diverse set of trajectory proposals, filters
these trajectories with a lightweight and interpretable safety filter, and then
uses a learned model to score each remaining trajectory. The best trajectory is
then tracked by the low-level controller of our self-driving vehicle. We train
our trajectory scoring model on a 500+ hour real-world dataset of expert
driving demonstrations in Las Vegas within the maximum entropy IRL framework.
DriveIRL's benefits include: a simple design due to only learning the
trajectory scoring function, relatively interpretable features, and strong
real-world performance. We validated DriveIRL on the Las Vegas Strip and
demonstrated fully autonomous driving in heavy traffic, including scenarios
involving cut-ins, abrupt braking by the lead vehicle, and hotel pickup/dropoff
zones. Our dataset will be made public to help further research in this area. |
2,627 | Distributionally Invariant Learning: Rationalization and Practical Algorithms | The invariance property across environments is at the heart of invariant
learning methods for the Out-of-Distribution (OOD) Generalization problem.
Although intuitively reasonable, strong assumptions on the availability and
quality of environments have to be made for the learnability of the strict
invariance property. Recently, to relax the requirements for environments
empirically, some works propose to learn pseudo-environments for invariant
learning. However, it could be misleading when pursuing strict invariance under
latent heterogeneity, since the underlying invariance could have been violated
during the pseudo-environment learning procedure. To this end, we come up with
the distributional invariance property as a relaxed alternative to the strict
invariance, which considers the invariance only among sub-populations down to a
prescribed scale and allows a certain degree of variation. We reformulate the
invariant learning problem under latent heterogeneity into a relaxed form that
pursues the distributional invariance, based on which we propose our novel
Distributionally Invariant Learning (DIL) framework as well as two
implementations named DIL-MMD and DIL-KL. Theoretically, we provide the
guarantees for the distributional invariance as well as bounds of the
generalization error gap. Extensive experimental results validate the
effectiveness of our proposed algorithms. |
2,628 | DynaMaR: Dynamic Prompt with Mask Token Representation | Recent research has shown that large language models pretrained using
unsupervised approaches can achieve significant performance improvement on many
downstream tasks. Typically when adapting these language models to downstream
tasks, like a classification or regression task, we employ a fine-tuning
paradigm in which the sentence representation from the language model is input
to a task-specific head; the model is then fine-tuned end-to-end. However, with
the emergence of models like GPT-3, prompt-based fine-tuning has been proven to
be a successful approach for few-shot tasks. Inspired by this work, we study
discrete prompt technologies in practice. There are two issues that arise with
the standard prompt approach. First, it can overfit on the prompt template.
Second, it requires manual effort to formulate the downstream task as a
language model problem. In this paper, we propose an improvement to
prompt-based fine-tuning that addresses these two issues. We refer to our
approach as DynaMaR -- Dynamic Prompt with Mask Token Representation. Results
show that DynaMaR can achieve an average improvement of 10% in few-shot
settings and improvement of 3.7% in data-rich settings over the standard
fine-tuning approach on four e-commerce applications. |
2,629 | DETR++: Taming Your Multi-Scale Detection Transformer | Convolutional Neural Networks (CNN) have dominated the field of detection
ever since the success of AlexNet in ImageNet classification [12]. With the
sweeping reform of Transformers [27] in natural language processing, Carion et
al. [2] introduce the Transformer-based detection method, i.e., DETR. However,
due to the quadratic complexity in the self-attention mechanism in the
Transformer, DETR is never able to incorporate multi-scale features as
performed in existing CNN-based detectors, leading to inferior results in small
object detection. To mitigate this issue and further improve performance of
DETR, in this work, we investigate different methods to incorporate multi-scale
features and find that a Bi-directional Feature Pyramid (BiFPN) works best with
DETR in further raising the detection precision. With this discovery, we
propose DETR++, a new architecture that improves detection results by 1.9% AP
on MS COCO 2017, 11.5% AP on RICO icon detection, and 9.1% AP on RICO layout
extraction over existing baselines. |
2,630 | Recall Distortion in Neural Network Pruning and the Undecayed Pruning Algorithm | Pruning techniques have been successfully used in neural networks to trade
accuracy for sparsity. However, the impact of network pruning is not uniform:
prior work has shown that the recall for underrepresented classes in a dataset
may be more negatively affected. In this work, we study such relative
distortions in recall by hypothesizing an intensification effect that is
inherent to the model. Namely, that pruning makes recall relatively worse for a
class with recall below accuracy and, conversely, that it makes recall
relatively better for a class with recall above accuracy. In addition, we
propose a new pruning algorithm aimed at attenuating such effect. Through
statistical analysis, we have observed that intensification is less severe with
our algorithm but nevertheless more pronounced with relatively more difficult
tasks, less complex models, and higher pruning ratios. More surprisingly, we
conversely observe a de-intensification effect with lower pruning ratios. |
2,631 | Decomposed Linear Dynamical Systems (dLDS) for learning the latent components of neural dynamics | Learning interpretable representations of neural dynamics at a population
level is a crucial first step to understanding how neural activity relates to
perception and behavior. Models of neural dynamics often focus on either
low-dimensional projections of neural activity, or on learning dynamical
systems that explicitly relate to the neural state over time. We discuss how
these two approaches are interrelated by considering dynamical systems as
representative of flows on a low-dimensional manifold. Building on this
concept, we propose a new decomposed dynamical system model that represents
complex non-stationary and nonlinear dynamics of time-series data as a sparse
combination of simpler, more interpretable components. The decomposed nature of
the dynamics generalizes over previous switched approaches and enables modeling
of overlapping and non-stationary drifts in the dynamics. We further present a
dictionary learning-driven approach to model fitting, where we leverage recent
results in tracking sparse vectors over time. We demonstrate that our model can
learn efficient representations and smooth transitions between dynamical modes
in both continuous-time and discrete-time examples. We show results on
low-dimensional linear and nonlinear attractors to demonstrate that our
decomposed dynamical systems model can well approximate nonlinear dynamics.
Additionally, we apply our model to C. elegans data, illustrating a diversity
of dynamics that is obscured when classified into discrete states. |
2,632 | A Simple and Optimal Policy Design with Safety against Heavy-tailed Risk for Multi-armed Bandits | We design new policies that ensure both worst-case optimality for expected
regret and light-tailed risk for regret distribution in the stochastic
multi-armed bandit problem. Recently, arXiv:2109.13595 showed that
information-theoretically optimized bandit algorithms suffer from some serious
heavy-tailed risk; that is, the worst-case probability of incurring a linear
regret slowly decays at a polynomial rate of $1/T$, as $T$ (the time horizon)
increases. Inspired by their results, we further show that widely used policies
(e.g., Upper Confidence Bound, Thompson Sampling) also incur heavy-tailed risk;
and this heavy-tailed risk actually exists for all "instance-dependent
consistent" policies. With the aim to ensure safety against such heavy-tailed
risk, starting from the two-armed bandit setting, we provide a simple policy
design that (i) has the worst-case optimality for the expected regret at order
$\tilde O(\sqrt{T})$ and (ii) has the worst-case tail probability of incurring
a linear regret decay at an optimal exponential rate $\exp(-\Omega(\sqrt{T}))$.
Next, we improve the policy design and analysis to the general $K$-armed bandit
setting. We provide explicit tail probability bound for any regret threshold
under our policy design. Specifically, the worst-case probability of incurring
a regret larger than $x$ is upper bounded by $\exp(-\Omega(x/\sqrt{KT}))$. We
also enhance the policy design to accommodate the "any-time" setting where $T$
is not known a priori, and prove equivalently desired policy performances as
compared to the "fixed-time" setting with known $T$. A brief account of
numerical experiments is conducted to illustrate the theoretical findings. Our
results reveal insights on the incompatibility between consistency and
light-tailed risk, whereas indicate that worst-case optimality on expected
regret and light-tailed risk on regret distribution are compatible. |
2,633 | Improving Knowledge Graph Embedding via Iterative Self-Semantic Knowledge Distillation | Knowledge graph embedding (KGE) has been intensively investigated for link
prediction by projecting entities and relations into continuous vector spaces.
Current popular high-dimensional KGE methods obtain quite slight performance
gains while require enormous computation and memory costs. In contrast to
high-dimensional KGE models, training low-dimensional models is more efficient
and worthwhile for better deployments to practical intelligent systems.
However, the model expressiveness of semantic information in knowledge graphs
(KGs) is highly limited in the low dimension parameter space. In this paper, we
propose iterative self-semantic knowledge distillation strategy to improve the
KGE model expressiveness in the low dimension space. KGE model combined with
our proposed strategy plays the teacher and student roles alternatively during
the whole training process. Specifically, at a certain iteration, the model is
regarded as a teacher to provide semantic information for the student. At next
iteration, the model is regard as a student to incorporate the semantic
information transferred from the teacher. We also design a novel semantic
extraction block to extract iteration-based semantic information for the
training model self-distillation. Iteratively incorporating and accumulating
iteration-based semantic information enables the low-dimensional model to be
more expressive for better link prediction in KGs. There is only one model
during the whole training, which alleviates the increase of computational
expensiveness and memory requirements. Furthermore, the proposed strategy is
model-agnostic and can be seamlessly combined with other KGE models. Consistent
and significant performance gains in experimental evaluations on four standard
datasets demonstrate the effectiveness of the proposed self-distillation
strategy. |
2,634 | Confounder Analysis in Measuring Representation in Product Funnels | This paper discusses an application of Shapley values in the causal inference
field, specifically on how to select the top confounder variables for coarsened
exact matching method in a scalable way. We use a dataset from an observational
experiment involving LinkedIn members as a use case to test its applicability,
and show that Shapley values are highly informational and can be leveraged for
its robust importance-ranking capability. |
2,635 | Beyond Faithfulness: A Framework to Characterize and Compare Saliency Methods | Saliency methods calculate how important each input feature is to a machine
learning model's prediction, and are commonly used to understand model
reasoning. "Faithfulness", or how fully and accurately the saliency output
reflects the underlying model, is an oft-cited desideratum for these methods.
However, explanation methods must necessarily sacrifice certain information in
service of user-oriented goals such as simplicity. To that end, and akin to
performance metrics, we frame saliency methods as abstractions: individual
tools that provide insight into specific aspects of model behavior and entail
tradeoffs. Using this framing, we describe a framework of nine dimensions to
characterize and compare the properties of saliency methods. We group these
dimensions into three categories that map to different phases of the
interpretation process: methodology, or how the saliency is calculated;
sensitivity, or relationships between the saliency result and the underlying
model or input; and, perceptibility, or how a user interprets the result. As we
show, these dimensions give us a granular vocabulary for describing and
comparing saliency methods -- for instance, allowing us to develop "saliency
cards" as a form of documentation, or helping downstream users understand
tradeoffs and choose a method for a particular use case. Moreover, by situating
existing saliency methods within this framework, we identify opportunities for
future work, including filling gaps in the landscape and developing new
evaluation metrics. |
2,636 | GRETEL: A unified framework for Graph Counterfactual Explanation Evaluation | Machine Learning (ML) systems are a building part of the modern tools which
impact our daily life in several application domains. Due to their black-box
nature, those systems are hardly adopted in application domains (e.g. health,
finance) where understanding the decision process is of paramount importance.
Explanation methods were developed to explain how the ML model has taken a
specific decision for a given case/instance. Graph Counterfactual Explanations
(GCE) is one of the explanation techniques adopted in the Graph Learning
domain. The existing works of Graph Counterfactual Explanations diverge mostly
in the problem definition, application domain, test data, and evaluation
metrics, and most existing works do not compare exhaustively against other
counterfactual explanation techniques present in the literature. We present
GRETEL, a unified framework to develop and test GCE methods in several
settings. GRETEL is a highly extensible evaluation framework which promotes the
Open Science and the evaluations reproducibility by providing a set of
well-defined mechanisms to integrate and manage easily: both real and synthetic
datasets, ML models, state-of-the-art explanation techniques, and evaluation
measures. To present GRETEL, we show the experiments conducted to integrate and
test several synthetic and real datasets with several existing explanation
techniques and base ML models. |
2,637 | Learning to Efficiently Propagate for Reasoning on Knowledge Graphs | Path-based methods are more appealing solutions than embedding methods for
knowledge graph reasoning, due to their interpretability and generalization
ability to unseen graphs. However, path-based methods usually suffer from the
problem of scalability, as the time complexity grows exponentially w.r.t. the
length of paths. While recent methods compute reasoning paths with the
Bellman-Ford algorithm in polynomial time, the time and memory cost remains
very high, as they need to propagate through all the nodes and edges in the
graph. In this paper, we propose A*Net, an efficient model for path-based
reasoning on knowledge graphs. Inspired by the classical A* algorithm for
shortest path problems, our A*Net prioritizes important nodes and edges at each
propagation step, to reduce the time and memory footprint. Unlike the classical
A* algorithm that uses a heuristic function, we propose to learn the priority
function for each node to capture the complex semantics in knowledge graphs.
The priority function and the propagation steps are jointly optimized through
backpropagation. Experiments on both transductive and inductive knowledge graph
reasoning benchmarks show that A*Net achieves competitive performance with
existing state-of-the-art path-based methods, and meanwhile reduces the number
of messages, the time and the memory cost up to 7.2$\times$, 3.4$\times$ and
4.9$\times$ respectively. |
2,638 | Robust Time Series Dissimilarity Measure for Outlier Detection and Periodicity Detection | Dynamic time warping (DTW) is an effective dissimilarity measure in many time
series applications. Despite its popularity, it is prone to noises and
outliers, which leads to singularity problem and bias in the measurement. The
time complexity of DTW is quadratic to the length of time series, making it
inapplicable in real-time applications. In this paper, we propose a novel time
series dissimilarity measure named RobustDTW to reduce the effects of noises
and outliers. Specifically, the RobustDTW estimates the trend and optimizes the
time warp in an alternating manner by utilizing our designed temporal graph
trend filtering. To improve efficiency, we propose a multi-level framework that
estimates the trend and the warp function at a lower resolution, and then
repeatedly refines them at a higher resolution. Based on the proposed
RobustDTW, we further extend it to periodicity detection and outlier time
series detection. Experiments on real-world datasets demonstrate the superior
performance of RobustDTW compared to DTW variants in both outlier time series
detection and periodicity detection. |
2,639 | Sampling without Replacement Leads to Faster Rates in Finite-Sum Minimax Optimization | We analyze the convergence rates of stochastic gradient algorithms for smooth
finite-sum minimax optimization and show that, for many such algorithms,
sampling the data points without replacement leads to faster convergence
compared to sampling with replacement. For the smooth and strongly
convex-strongly concave setting, we consider gradient descent ascent and the
proximal point method, and present a unified analysis of two popular
without-replacement sampling strategies, namely Random Reshuffling (RR), which
shuffles the data every epoch, and Single Shuffling or Shuffle Once (SO), which
shuffles only at the beginning. We obtain tight convergence rates for RR and SO
and demonstrate that these strategies lead to faster convergence than uniform
sampling. Moving beyond convexity, we obtain similar results for smooth
nonconvex-nonconcave objectives satisfying a two-sided Polyak-{\L}ojasiewicz
inequality. Finally, we demonstrate that our techniques are general enough to
analyze the effect of data-ordering attacks, where an adversary manipulates the
order in which data points are supplied to the optimizer. Our analysis also
recovers tight rates for the incremental gradient method, where the data points
are not shuffled at all. |
2,640 | On the Convergence of Optimizing Persistent-Homology-Based Losses | Topological loss based on persistent homology has shown promise in various
applications. A topological loss enforces the model to achieve certain desired
topological property. Despite its empirical success, less is known about the
optimization behavior of the loss. In fact, the topological loss involves
combinatorial configurations that may oscillate during optimization. In this
paper, we introduce a general purpose regularized topology-aware loss. We
propose a novel regularization term and also modify existing topological loss.
These contributions lead to a new loss function that not only enforces the
model to have desired topological behavior, but also achieves satisfying
convergence behavior. Our main theoretical result guarantees that the loss can
be optimized efficiently, under mild assumptions. |
2,641 | Predicting Electricity Infrastructure Induced Wildfire Risk in California | This paper examines the use of risk models to predict the timing and location
of wildfires caused by electricity infrastructure. Our data include historical
ignition and wire-down points triggered by grid infrastructure collected
between 2015 to 2019 in Pacific Gas & Electricity territory along with various
weather, vegetation, and very high resolution data on grid infrastructure
including location, age, materials. With these data we explore a range of
machine learning methods and strategies to manage training data imbalance. The
best area under the receiver operating characteristic we obtain is 0.776 for
distribution feeder ignitions and 0.824 for transmission line wire-down events,
both using the histogram-based gradient boosting tree algorithm (HGB) with
under-sampling. We then use these models to identify which information provides
the most predictive value. After line length, we find that weather and
vegetation features dominate the list of top important features for ignition or
wire-down risk. Distribution ignition models show more dependence on
slow-varying vegetation variables such as burn index, energy release content,
and tree height, whereas transmission wire-down models rely more on primary
weather variables such as wind speed and precipitation. These results point to
the importance of improved vegetation modeling for feeder ignition risk models,
and improved weather forecasting for transmission wire-down models. We observe
that infrastructure features make small but meaningful improvements to risk
model predictive power. |
2,642 | Neuro-Symbolic Causal Language Planning with Commonsense Prompting | Language planning aims to implement complex high-level goals by decomposition
into sequential simpler low-level steps. Such procedural reasoning ability is
essential for applications such as household robots and virtual assistants.
Although language planning is a basic skill set for humans in daily life, it
remains a challenge for large language models (LLMs) that lack deep-level
commonsense knowledge in the real world. Previous methods require either manual
exemplars or annotated programs to acquire such ability from LLMs. In contrast,
this paper proposes Neuro-Symbolic Causal Language Planner (CLAP) that elicits
procedural knowledge from the LLMs with commonsense-infused prompting.
Pre-trained knowledge in LLMs is essentially an unobserved confounder that
causes spurious correlations between tasks and action plans. Through the lens
of a Structural Causal Model (SCM), we propose an effective strategy in CLAP to
construct prompts as a causal intervention toward our SCM. Using graph sampling
techniques and symbolic program executors, our strategy formalizes the
structured causal prompts from commonsense knowledge bases. CLAP obtains
state-of-the-art performance on WikiHow and RobotHow, achieving a relative
improvement of 5.28% in human evaluations under the counterfactual setting.
This indicates the superiority of CLAP in causal language planning semantically
and sequentially. |
2,643 | Spectral Bias Outside the Training Set for Deep Networks in the Kernel Regime | We provide quantitative bounds measuring the $L^2$ difference in function
space between the trajectory of a finite-width network trained on finitely many
samples from the idealized kernel dynamics of infinite width and infinite data.
An implication of the bounds is that the network is biased to learn the top
eigenfunctions of the Neural Tangent Kernel not just on the training set but
over the entire input space. This bias depends on the model architecture and
input distribution alone and thus does not depend on the target function which
does not need to be in the RKHS of the kernel. The result is valid for deep
architectures with fully connected, convolutional, and residual layers.
Furthermore the width does not need to grow polynomially with the number of
samples in order to obtain high probability bounds up to a stopping time. The
proof exploits the low-effective-rank property of the Fisher Information Matrix
at initialization, which implies a low effective dimension of the model (far
smaller than the number of parameters). We conclude that local capacity control
from the low effective rank of the Fisher Information Matrix is still
underexplored theoretically. |
2,644 | Tight basis cycle representatives for persistent homology of large data sets | Persistent homology (PH) is a popular tool for topological data analysis that
has found applications across diverse areas of research. It provides a rigorous
method to compute robust topological features in discrete experimental
observations that often contain various sources of uncertainties. Although
powerful in theory, PH suffers from high computation cost that precludes its
application to large data sets. Additionally, most analyses using PH are
limited to computing the existence of nontrivial features. Precise localization
of these features is not generally attempted because, by definition, localized
representations are not unique and because of even higher computation cost. For
scientific applications, such a precise location is a sine qua non for
determining functional significance. Here, we provide a strategy and algorithms
to compute tight representative boundaries around nontrivial robust features in
large data sets. To showcase the efficiency of our algorithms and the precision
of computed boundaries, we analyze three data sets from different scientific
fields. In the human genome, we found an unexpected effect on loops through
chromosome 13 and the sex chromosomes, upon impairment of chromatin loop
formation. In a distribution of galaxies in the universe, we found
statistically significant voids. In protein homologs with significantly
different topology, we found voids attributable to ligand-interaction,
mutation, and differences between species. |
2,645 | Stable and memory-efficient image recovery using monotone operator learning (MOL) | We introduce a monotone deep equilibrium learning framework for large-scale
inverse problems in imaging. The proposed algorithm relies on forward-backward
splitting, where each iteration consists of a gradient descent involving the
score function and a conjugate gradient algorithm to encourage data
consistency. The score function is modeled as a monotone convolutional neural
network. The use of a monotone operator offers several benefits, including
guaranteed convergence, uniqueness of fixed point, and robustness to input
perturbations, similar to the use of convex priors in compressive sensing. In
addition, the proposed formulation is significantly more memory-efficient than
unrolled methods, which allows us to apply it to 3D problems that current
unrolled algorithms cannot handle. Experiments show that the proposed scheme
can offer improved performance in 3D settings while being stable in the
presence of input perturbations. |
2,646 | Schema-Guided Event Graph Completion | We tackle a new task, event graph completion, which aims to predict missing
event nodes for event graphs. Existing link prediction or graph completion
methods have difficulty dealing with event graphs because they are usually
designed for a single large graph such as a social network or a knowledge
graph, rather than multiple small dynamic event graphs. Moreover, they can only
predict missing edges rather than missing nodes. In this work, we propose to
utilize event schema, a template that describes the stereotypical structure of
event graphs, to address the above issues. Our schema-guided event graph
completion approach first maps an instance event graph to a subgraph of the
schema graph by a heuristic subgraph matching algorithm. Then it predicts
whether a candidate event node in the schema graph should be added to the
instantiated schema subgraph by characterizing two types of local topology of
the schema graph: neighbors of the candidate node and the subgraph, and paths
that connect the candidate node and the subgraph. These two modules are later
combined together for the final prediction. We also propose a self-supervised
strategy to construct training samples, as well as an inference algorithm that
is specifically designed to complete event graphs. Extensive experimental
results on four datasets demonstrate that our proposed method achieves
state-of-the-art performance, with 4.3% to 19.4% absolute F1 gains over the
best baseline method on the four datasets. |
2,647 | Remember the Past: Distilling Datasets into Addressable Memories for Neural Networks | We propose an algorithm that compresses the critical information of a large
dataset into compact addressable memories. These memories can then be recalled
to quickly re-train a neural network and recover the performance (instead of
storing and re-training on the full original dataset).
Building upon the dataset distillation framework, we make a key observation
that a shared common representation allows for more efficient and effective
distillation. Concretely, we learn a set of bases (aka "memories") which are
shared between classes and combined through learned flexible addressing
functions to generate a diverse set of training examples. This leads to several
benefits: 1) the size of compressed data does not necessarily grow linearly
with the number of classes; 2) an overall higher compression rate with more
effective distillation is achieved; and 3) more generalized queries are allowed
beyond recalling the original classes.
We demonstrate state-of-the-art results on the dataset distillation task
across five benchmarks, including up to 16.5% and 9.7% in retained accuracy
improvement when distilling CIFAR10 and CIFAR100 respectively. We then leverage
our framework to perform continual learning, achieving state-of-the-art results
on four benchmarks, with 23.2% accuracy improvement on MANY. |
2,648 | Training Subset Selection for Weak Supervision | Existing weak supervision approaches use all the data covered by weak signals
to train a classifier. We show both theoretically and empirically that this is
not always optimal. Intuitively, there is a tradeoff between the amount of
weakly-labeled data and the precision of the weak labels. We explore this
tradeoff by combining pretrained data representations with the cut statistic
(Muhlenbach et al., 2004) to select (hopefully) high-quality subsets of the
weakly-labeled training data. Subset selection applies to any label model and
classifier and is very simple to plug in to existing weak supervision
pipelines, requiring just a few lines of code. We show our subset selection
method improves the performance of weak supervision for a wide range of label
models, classifiers, and datasets. Using less weakly-labeled data improves the
accuracy of weak supervision pipelines by up to 19% (absolute) on benchmark
tasks. |
2,649 | 8-bit Numerical Formats for Deep Neural Networks | Given the current trend of increasing size and complexity of machine learning
architectures, it has become of critical importance to identify new approaches
to improve the computational efficiency of model training. In this context, we
address the advantages of floating-point over fixed-point representation, and
present an in-depth study on the use of 8-bit floating-point number formats for
activations, weights, and gradients for both training and inference. We explore
the effect of different bit-widths for exponents and significands and different
exponent biases. The experimental results demonstrate that a suitable choice of
these low-precision formats enables faster training and reduced power
consumption without any degradation in accuracy for a range of deep learning
models for image classification and language processing. |
2,650 | Boundary informed inverse PDE problems on discrete Riemann surfaces | We employ neural networks to tackle inverse partial differential equations on
discretized Riemann surfaces with boundary. To this end, we introduce the
concept of a graph with boundary which models these surfaces in a natural way.
Our method uses a message passing technique to keep track of an unknown
differential operator while using neural ODE solvers through the method of
lines to capture the evolution in time. As training data, we use noisy and
incomplete observations of sheaves on graphs at various timestamps. The novelty
of this approach is in working with manifolds with nontrivial topology and
utilizing the data on the graph boundary through a teacher forcing technique.
Despite the increasing interest in learning dynamical systems from finite
observations, many current methods are limited in two general ways: first, they
work with topologically trivial spaces, and second, they fail to handle the
boundary data on the ground space in a systematic way. The present work is an
attempt at addressing these limitations. We run experiments with synthetic data
of linear and nonlinear diffusion systems on orientable surfaces with positive
genus and boundary, and moreover, provide evidences for improvements upon the
existing paradigms. |
2,651 | Self-supervised Learning for Human Activity Recognition Using 700,000 Person-days of Wearable Data | Advances in deep learning for human activity recognition have been relatively
limited due to the lack of large labelled datasets. In this study, we leverage
self-supervised learning techniques on the UK-Biobank activity tracker
dataset--the largest of its kind to date--containing more than 700,000
person-days of unlabelled wearable sensor data. Our resulting activity
recognition model consistently outperformed strong baselines across seven
benchmark datasets, with an F1 relative improvement of 2.5%-100% (median
18.4%), the largest improvements occurring in the smaller datasets. In contrast
to previous studies, our results generalise across external datasets, devices,
and environments. Our open-source model will help researchers and developers to
build customisable and generalisable activity classifiers with high
performance. |
2,652 | Goal-Space Planning with Subgoal Models | This paper investigates a new approach to model-based reinforcement learning
using background planning: mixing (approximate) dynamic programming updates and
model-free updates, similar to the Dyna architecture. Background planning with
learned models is often worse than model-free alternatives, such as Double DQN,
even though the former uses significantly more memory and computation. The
fundamental problem is that learned models can be inaccurate and often generate
invalid states, especially when iterated many steps. In this paper, we avoid
this limitation by constraining background planning to a set of (abstract)
subgoals and learning only local, subgoal-conditioned models. This goal-space
planning (GSP) approach is more computationally efficient, naturally
incorporates temporal abstraction for faster long-horizon planning and avoids
learning the transition dynamics entirely. We show that our GSP algorithm can
learn significantly faster than a Double DQN baseline in a variety of
situations. |
2,653 | Distributive Justice as the Foundational Premise of Fair ML: Unification, Extension, and Interpretation of Group Fairness Metrics | Group fairness metrics are an established way of assessing the fairness of
prediction-based decision-making systems. However, these metrics are still
insufficiently linked to philosophical theories, and their moral meaning is
often unclear. We propose a general framework for analyzing the fairness of
decision systems based on theories of distributive justice, encompassing
different established "patterns of justice" that correspond to different
normative positions. We show that the most popular group fairness metrics can
be interpreted as special cases of our approach. Thus, we provide a unifying
and interpretative framework for group fairness metrics that reveals the
normative choices associated with each of them and that allows understanding
their moral substance. At the same time, we provide an extension of the space
of possible fairness metrics beyond the ones currently discussed in the fair ML
literature. Our framework also allows overcoming several limitations of group
fairness metrics that have been criticized in the literature, most notably (1)
that they are parity-based, i.e., that they demand some form of equality
between groups, which may sometimes be harmful to marginalized groups, (2) that
they only compare decisions across groups, but not the resulting consequences
for these groups, and (3) that the full breadth of the distributive justice
literature is not sufficiently represented. |
2,654 | A Justice-Based Framework for the Analysis of Algorithmic Fairness-Utility Trade-Offs | In prediction-based decision-making systems, different perspectives can be at
odds: The short-term business goals of the decision makers are often in
conflict with the decision subjects' wish to be treated fairly. Balancing these
two perspectives is a question of values. We provide a framework to make these
value-laden choices clearly visible. For this, we assume that we are given a
trained model and want to find decision rules that balance the perspective of
the decision maker and of the decision subjects. We provide an approach to
formalize both perspectives, i.e., to assess the utility of the decision maker
and the fairness towards the decision subjects. In both cases, the idea is to
elicit values from decision makers and decision subjects that are then turned
into something measurable. For the fairness evaluation, we build on the
literature on welfare-based fairness and ask what a fair distribution of
utility (or welfare) looks like. In this step, we build on well-known theories
of distributive justice. This allows us to derive a fairness score that we then
compare to the decision maker's utility for many different decision rules. This
way, we provide an approach for balancing the utility of the decision maker and
the fairness towards the decision subjects for a prediction-based
decision-making system. |
2,655 | Conditional Seq2Seq model for the time-dependent two-level system | We apply the deep learning neural network architecture to the two-level
system in quantum optics to solve the time-dependent Schrodinger equation. By
carefully designing the network structure and tuning parameters, above 90
percent accuracy in super long-term predictions can be achieved in the case of
random electric fields, which indicates a promising new method to solve the
time-dependent equation for two-level systems. By slightly modifying this
network, we think that this method can solve the two- or three-dimensional
time-dependent Schrodinger equation more efficiently than traditional
approaches. |
2,656 | Sample Complexity of Nonparametric Off-Policy Evaluation on Low-Dimensional Manifolds using Deep Networks | We consider the off-policy evaluation problem of reinforcement learning using
deep neural networks. We analyze the deep fitted Q-evaluation method for
estimating the expected cumulative reward of a target policy, when the data are
generated from an unknown behavior policy. We show that, by choosing network
size appropriately, one can leverage the low-dimensional manifold structure in
the Markov decision process and obtain a sample-efficient estimator without
suffering from the curse of high representation dimensionality. Specifically,
we establish a sharp error bound for the fitted Q-evaluation that depends on
the intrinsic low dimension, the smoothness of the state-action space, and a
function class-restricted $\chi^2$-divergence. It is noteworthy that the
restricted $\chi^2$-divergence measures the behavior and target policies' {\it
mismatch in the function space}, which can be small even if the two policies
are not close to each other in their tabular forms. Numerical experiments are
provided to support our theoretical analysis. |
2,657 | Graph Rationalization with Environment-based Augmentations | Rationale is defined as a subset of input features that best explains or
supports the prediction by machine learning models. Rationale identification
has improved the generalizability and interpretability of neural networks on
vision and language data. In graph applications such as molecule and polymer
property prediction, identifying representative subgraph structures named as
graph rationales plays an essential role in the performance of graph neural
networks. Existing graph pooling and/or distribution intervention methods
suffer from lack of examples to learn to identify optimal graph rationales. In
this work, we introduce a new augmentation operation called environment
replacement that automatically creates virtual data examples to improve
rationale identification. We propose an efficient framework that performs
rationale-environment separation and representation learning on the real and
augmented examples in latent spaces to avoid the high complexity of explicit
graph decoding and encoding. Comparing against recent techniques, experiments
on seven molecular and four polymer real datasets demonstrate the effectiveness
and efficiency of the proposed augmentation-based graph rationalization
framework. |
2,658 | A Human-Centric Take on Model Monitoring | Predictive models are increasingly used to make various consequential
decisions in high-stakes domains such as healthcare, finance, and policy. It
becomes critical to ensure that these models make accurate predictions, are
robust to shifts in the data, do not rely on spurious features, and do not
unduly discriminate against minority groups. To this end, several approaches
spanning various areas such as explainability, fairness, and robustness have
been proposed in recent literature. Such approaches need to be human-centered
as they cater to the understanding of the models to their users. However, there
is a research gap in understanding the human-centric needs and challenges of
monitoring machine learning (ML) models once they are deployed. To fill this
gap, we conducted an interview study with 13 practitioners who have experience
at the intersection of deploying ML models and engaging with customers spanning
domains such as financial services, healthcare, hiring, online retail,
computational advertising, and conversational assistants. We identified various
human-centric challenges and requirements for model monitoring in real-world
applications. Specifically, we found the need and the challenge for the model
monitoring systems to clarify the impact of the monitoring observations on
outcomes. Further, such insights must be actionable, robust, customizable for
domain-specific use cases, and cognitively considerate to avoid information
overload. |
2,659 | Efficient entity-based reinforcement learning | Recent deep reinforcement learning (DRL) successes rely on end-to-end
learning from fixed-size observational inputs (e.g. image, state-variables).
However, many challenging and interesting problems in decision making involve
observations or intermediary representations which are best described as a set
of entities: either the image-based approach would miss small but important
details in the observations (e.g. ojects on a radar, vehicles on satellite
images, etc.), the number of sensed objects is not fixed (e.g. robotic
manipulation), or the problem simply cannot be represented in a meaningful way
as an image (e.g. power grid control, or logistics). This type of structured
representations is not directly compatible with current DRL architectures,
however, there has been an increase in machine learning techniques directly
targeting structured information, potentially addressing this issue. We propose
to combine recent advances in set representations with slot attention and graph
neural networks to process structured data, broadening the range of
applications of DRL algorithms. This approach allows to address entity-based
problems in an efficient and scalable way. We show that it can improve training
time and robustness significantly, and demonstrate their potential to handle
structured as well as purely visual domains, on multiple environments from the
Atari Learning Environment and Simple Playgrounds. |
2,660 | A Bird's-Eye Tutorial of Graph Attention Architectures | Graph Neural Networks (GNNs) have shown tremendous strides in performance for
graph-structured problems especially in the domains of natural language
processing, computer vision and recommender systems. Inspired by the success of
the transformer architecture, there has been an ever-growing body of work on
attention variants of GNNs attempting to advance the state of the art in many
of these problems. Incorporating "attention" into graph mining has been viewed
as a way to overcome the noisiness, heterogenity and complexity associated with
graph-structured data as well as to encode soft-inductive bias. It is hence
crucial and advantageous to study these variants from a bird's-eye view to
assess their strengths and weaknesses. We provide a systematic and focused
tutorial centered around attention based GNNs in a hope to benefit researchers
dealing with graph-structured problems. Our tutorial looks at GNN variants from
the point of view of the attention function and iteratively builds the reader's
understanding of different graph attention variants. |
2,661 | On Efficient Approximate Queries over Machine Learning Models | The question of answering queries over ML predictions has been gaining
attention in the database community. This question is challenging because the
cost of finding high quality answers corresponds to invoking an oracle such as
a human expert or an expensive deep neural network model on every single item
in the DB and then applying the query. We develop a novel unified framework for
approximate query answering by leveraging a proxy to minimize the oracle usage
of finding high quality answers for both Precision-Target (PT) and
Recall-Target (RT) queries. Our framework uses a judicious combination of
invoking the expensive oracle on data samples and applying the cheap proxy on
the objects in the DB. It relies on two assumptions. Under the Proxy Quality
assumption, proxy quality can be quantified in a probabilistic manner w.r.t.
the oracle. This allows us to develop two algorithms: PQA that efficiently
finds high quality answers with high probability and no oracle calls, and PQE,
a heuristic extension that achieves empirically good performance with a small
number of oracle calls. Alternatively, under the Core Set Closure assumption,
we develop two algorithms: CSC that efficiently returns high quality answers
with high probability and minimal oracle usage, and CSE, which extends it to
more general settings. Our extensive experiments on five real-world datasets on
both query types, PT and RT, demonstrate that our algorithms outperform the
state-of-the-art and achieve high result quality with provable statistical
guarantees. |
2,662 | Invertible Sharpening Network for MRI Reconstruction Enhancement | High-quality MRI reconstruction plays a critical role in clinical
applications. Deep learning-based methods have achieved promising results on
MRI reconstruction. However, most state-of-the-art methods were designed to
optimize the evaluation metrics commonly used for natural images, such as PSNR
and SSIM, whereas the visual quality is not primarily pursued. Compared to the
fully-sampled images, the reconstructed images are often blurry, where
high-frequency features might not be sharp enough for confident clinical
diagnosis. To this end, we propose an invertible sharpening network
(InvSharpNet) to improve the visual quality of MRI reconstructions. During
training, unlike the traditional methods that learn to map the input data to
the ground truth, InvSharpNet adapts a backward training strategy that learns a
blurring transform from the ground truth (fully-sampled image) to the input
data (blurry reconstruction). During inference, the learned blurring transform
can be inverted to a sharpening transform leveraging the network's
invertibility. The experiments on various MRI datasets demonstrate that
InvSharpNet can improve reconstruction sharpness with few artifacts. The
results were also evaluated by radiologists, indicating better visual quality
and diagnostic confidence of our proposed method. |
2,663 | Collaborative Linear Bandits with Adversarial Agents: Near-Optimal Regret Bounds | We consider a linear stochastic bandit problem involving $M$ agents that can
collaborate via a central server to minimize regret. A fraction $\alpha$ of
these agents are adversarial and can act arbitrarily, leading to the following
tension: while collaboration can potentially reduce regret, it can also disrupt
the process of learning due to adversaries. In this work, we provide a
fundamental understanding of this tension by designing new algorithms that
balance the exploration-exploitation trade-off via carefully constructed robust
confidence intervals. We also complement our algorithms with tight analyses.
First, we develop a robust collaborative phased elimination algorithm that
achieves $\tilde{O}\left(\alpha+ 1/\sqrt{M}\right) \sqrt{dT}$ regret for each
good agent; here, $d$ is the model-dimension and $T$ is the horizon. For small
$\alpha$, our result thus reveals a clear benefit of collaboration despite
adversaries. Using an information-theoretic argument, we then prove a matching
lower bound, thereby providing the first set of tight, near-optimal regret
bounds for collaborative linear bandits with adversaries. Furthermore, by
leveraging recent advances in high-dimensional robust statistics, we
significantly extend our algorithmic ideas and results to (i) the generalized
linear bandit model that allows for non-linear observation maps; and (ii) the
contextual bandit setting that allows for time-varying feature vectors. |
2,664 | RORL: Robust Offline Reinforcement Learning via Conservative Smoothing | Offline reinforcement learning (RL) provides a promising direction to exploit
the massive amount of offline data for complex decision-making tasks. Due to
the distribution shift issue, current offline RL algorithms are generally
designed to be conservative for value estimation and action selection. However,
such conservatism impairs the robustness of learned policies, leading to a
significant change even for a small perturbation on observations. To trade off
robustness and conservatism, we propose Robust Offline Reinforcement Learning
(RORL) with a novel conservative smoothing technique. In RORL, we explicitly
introduce regularization on the policy and the value function for states near
the dataset and additional conservative value estimation on these OOD states.
Theoretically, we show RORL enjoys a tighter suboptimality bound than recent
theoretical results in linear MDPs. We demonstrate that RORL can achieve the
state-of-the-art performance on the general offline RL benchmark and is
considerably robust to adversarial observation perturbation. |
2,665 | Deep Learning Models of the Discrete Component of the Galactic Interstellar Gamma-Ray Emission | A significant point-like component from the small scale (or discrete)
structure in the H2 interstellar gas might be present in the Fermi-LAT data,
but modeling this emission relies on observations of rare gas tracers only
available in limited regions of the sky. Identifying this contribution is
important to discriminate gamma-ray point sources from interstellar gas, and to
better characterize extended gamma-ray sources. We design and train
convolutional neural networks to predict this emission where observations of
these rare tracers do not exist and discuss the impact of this component on the
analysis of the Fermi-LAT data. In particular, we evaluate prospects to exploit
this methodology in the characterization of the Fermi-LAT Galactic center
excess through accurate modeling of point-like structures in the data to help
distinguish between a point-like or smooth nature for the excess. We show that
deep learning may be effectively employed to model the gamma-ray emission
traced by these rare H2 proxies within statistical significance in data-rich
regions, supporting prospects to employ these methods in yet unobserved
regions. |
2,666 | Quantum Neural Network Classifiers: A Tutorial | Machine learning has achieved dramatic success over the past decade, with
applications ranging from face recognition to natural language processing.
Meanwhile, rapid progress has been made in the field of quantum computation
including developing both powerful quantum algorithms and advanced quantum
devices. The interplay between machine learning and quantum physics holds the
intriguing potential for bringing practical applications to the modern society.
Here, we focus on quantum neural networks in the form of parameterized quantum
circuits. We will mainly discuss different structures and encoding strategies
of quantum neural networks for supervised learning tasks, and benchmark their
performance utilizing Yao.jl, a quantum simulation package written in Julia
Language. The codes are efficient, aiming to provide convenience for beginners
in scientific works such as developing powerful variational quantum learning
models and assisting the corresponding experimental demonstrations. |
2,667 | GenSDF: Two-Stage Learning of Generalizable Signed Distance Functions | We investigate the generalization capabilities of neural signed distance
functions (SDFs) for learning 3D object representations for unseen and
unlabeled point clouds. Existing methods can fit SDFs to a handful of object
classes and boast fine detail or fast inference speeds, but do not generalize
well to unseen shapes. We introduce a two-stage semi-supervised meta-learning
approach that transfers shape priors from labeled to unlabeled data to
reconstruct unseen object categories. The first stage uses an episodic training
scheme to simulate training on unlabeled data and meta-learns initial shape
priors. The second stage then introduces unlabeled data with disjoint classes
in a semi-supervised scheme to diversify these priors and achieve
generalization. We assess our method on both synthetic data and real collected
point clouds. Experimental results and analysis validate that our approach
outperforms existing neural SDF methods and is capable of robust zero-shot
inference on 100+ unseen classes. Code can be found at
https://github.com/princeton-computational-imaging/gensdf. |
2,668 | Blended Latent Diffusion | The tremendous progress in neural image generation, coupled with the
emergence of seemingly omnipotent vision-language models has finally enabled
text-based interfaces for creating and editing images. Handling generic images
requires a diverse underlying generative model, hence the latest works utilize
diffusion models, which were shown to surpass GANs in terms of diversity. One
major drawback of diffusion models, however, is their relatively slow inference
time. In this paper, we present an accelerated solution to the task of local
text-driven editing of generic images, where the desired edits are confined to
a user-provided mask. Our solution leverages a recent text-to-image Latent
Diffusion Model (LDM), which speeds up diffusion by operating in a
lower-dimensional latent space. We first convert the LDM into a local image
editor by incorporating Blended Diffusion into it. Next we propose an
optimization-based solution for the inherent inability of this LDM to
accurately reconstruct images. Finally, we address the scenario of performing
local edits using thin masks. We evaluate our method against the available
baselines both qualitatively and quantitatively and demonstrate that in
addition to being faster, our method achieves better precision than the
baselines while mitigating some of their artifacts. Project page is available
at https://omriavrahami.com/blended-latent-diffusion-page/ |
2,669 | Neuro CROSS exchange: Learning to CROSS exchange to solve realistic vehicle routing problems | CROSS exchange (CE), a meta-heuristic that solves various vehicle routing
problems (VRPs), improves the solutions of VRPs by swapping the sub-tours of
the vehicles. Inspired by CE, we propose Neuro CE (NCE), a fundamental operator
of learned meta-heuristic, to solve various VRPs while overcoming the
limitations of CE (i.e., the expensive $\mathcal{O}(n^4)$ search cost). NCE
employs a graph neural network to predict the cost-decrements (i.e., results of
CE searches) and utilizes the predicted cost-decrements as guidance for search
to decrease the search cost to $\mathcal{O}(n^2)$. As the learning objective of
NCE is to predict the cost-decrement, the training can be simply done in a
supervised fashion, whose training samples can be prepared effortlessly.
Despite the simplicity of NCE, numerical results show that the NCE trained with
flexible multi-depot VRP (FMDVRP) outperforms the meta-heuristic baselines.
More importantly, it significantly outperforms the neural baselines when
solving distinctive special cases of FMDVRP (e.g., MDVRP, mTSP, CVRP) without
additional training. |
2,670 | The Neural Covariance SDE: Shaped Infinite Depth-and-Width Networks at Initialization | The logit outputs of a feedforward neural network at initialization are
conditionally Gaussian, given a random covariance matrix defined by the
penultimate layer. In this work, we study the distribution of this random
matrix. Recent work has shown that shaping the activation function as network
depth grows large is necessary for this covariance matrix to be non-degenerate.
However, the current infinite-width-style understanding of this shaping method
is unsatisfactory for large depth: infinite-width analyses ignore the
microscopic fluctuations from layer to layer, but these fluctuations accumulate
over many layers.
To overcome this shortcoming, we study the random covariance matrix in the
shaped infinite-depth-and-width limit. We identify the precise scaling of the
activation function necessary to arrive at a non-trivial limit, and show that
the random covariance matrix is governed by a stochastic differential equation
(SDE) that we call the Neural Covariance SDE. Using simulations, we show that
the SDE closely matches the distribution of the random covariance matrix of
finite networks. Additionally, we recover an if-and-only-if condition for
exploding and vanishing norms of large shaped networks based on the activation
function. |
2,671 | Communication-constrained hypothesis testing: Optimality, robustness, and reverse data processing inequalities | We study hypothesis testing under communication constraints, where each
sample is quantized before being revealed to a statistician. Without
communication constraints, it is well known that the sample complexity of
simple binary hypothesis testing is characterized by the Hellinger distance
between the distributions. We show that the sample complexity of simple binary
hypothesis testing under communication constraints is at most a logarithmic
factor larger than in the unconstrained setting and this bound is tight. We
develop a polynomial-time algorithm that achieves the aforementioned sample
complexity. Our framework extends to robust hypothesis testing, where the
distributions are corrupted in the total variation distance. Our proofs rely on
a new reverse data processing inequality and a reverse Markov inequality, which
may be of independent interest. For simple $M$-ary hypothesis testing, the
sample complexity in the absence of communication constraints has a logarithmic
dependence on $M$. We show that communication constraints can cause an
exponential blow-up leading to $\Omega(M)$ sample complexity even for adaptive
algorithms. |
2,672 | Dual Decomposition of Convex Optimization Layers for Consistent Attention in Medical Images | A key concern in integrating machine learning models in medicine is the
ability to interpret their reasoning. Popular explainability methods have
demonstrated satisfactory results in natural image recognition, yet in medical
image analysis, many of these approaches provide partial and noisy
explanations. Recently, attention mechanisms have shown compelling results both
in their predictive performance and in their interpretable qualities. A
fundamental trait of attention is that it leverages salient parts of the input
which contribute to the model's prediction. To this end, our work focuses on
the explanatory value of attention weight distributions. We propose a
multi-layer attention mechanism that enforces consistent interpretations
between attended convolutional layers using convex optimization. We apply
duality to decompose the consistency constraints between the layers by
reparameterizing their attention probability distributions. We further suggest
learning the dual witness by optimizing with respect to our objective; thus,
our implementation uses standard back-propagation, hence it is highly
efficient. While preserving predictive performance, our proposed method
leverages weakly annotated medical imaging data and provides complete and
faithful explanations to the model's prediction. |
2,673 | Robust Calibration with Multi-domain Temperature Scaling | Uncertainty quantification is essential for the reliable deployment of
machine learning models to high-stakes application domains. Uncertainty
quantification is all the more challenging when training distribution and test
distribution are different, even the distribution shifts are mild. Despite the
ubiquity of distribution shifts in real-world applications, existing
uncertainty quantification approaches mainly study the in-distribution setting
where the train and test distributions are the same. In this paper, we develop
a systematic calibration model to handle distribution shifts by leveraging data
from multiple domains. Our proposed method -- multi-domain temperature scaling
-- uses the heterogeneity in the domains to improve calibration robustness
under distribution shift. Through experiments on three benchmark data sets, we
find our proposed method outperforms existing methods as measured on both
in-distribution and out-of-distribution test sets. |
2,674 | Predicting and Understanding Human Action Decisions during Skillful Joint-Action via Machine Learning and Explainable-AI | This study uses supervised machine learning (SML) and explainable artificial
intelligence (AI) to model, predict and understand human decision-making during
skillful joint-action. Long short-term memory networks were trained to predict
the target selection decisions of expert and novice actors completing a dyadic
herding task. Results revealed that the trained models were expertise specific
and could not only accurately predict the target selection decisions of expert
and novice herders but could do so at timescales that preceded an actor's
conscious intent. To understand what differentiated the target selection
decisions of expert and novice actors, we then employed the explainable-AI
technique, SHapley Additive exPlanation, to identify the importance of
informational features (variables) on model predictions. This analysis revealed
that experts were more influenced by information about the state of their
co-herders compared to novices. The utility of employing SML and explainable-AI
techniques for investigating human decision-making is discussed. |
2,675 | Global Mixup: Eliminating Ambiguity with Clustering | Data augmentation with \textbf{Mixup} has been proven an effective method to
regularize the current deep neural networks. Mixup generates virtual samples
and corresponding labels at once through linear interpolation. However, this
one-stage generation paradigm and the use of linear interpolation have the
following two defects: (1) The label of the generated sample is directly
combined from the labels of the original sample pairs without reasonable
judgment, which makes the labels likely to be ambiguous. (2) linear combination
significantly limits the sampling space for generating samples. To tackle these
problems, we propose a novel and effective augmentation method based on global
clustering relationships named \textbf{Global Mixup}. Specifically, we
transform the previous one-stage augmentation process into two-stage,
decoupling the process of generating virtual samples from the labeling. And for
the labels of the generated samples, relabeling is performed based on
clustering by calculating the global relationships of the generated samples. In
addition, we are no longer limited to linear relationships but generate more
reliable virtual samples in a larger sampling space. Extensive experiments for
\textbf{CNN}, \textbf{LSTM}, and \textbf{BERT} on five tasks show that Global
Mixup significantly outperforms previous state-of-the-art baselines. Further
experiments also demonstrate the advantage of Global Mixup in low-resource
scenarios. |
2,676 | Robust and Fast Data-Driven Power System State Estimator Using Graph Neural Networks | The power system state estimation (SE) algorithm estimates the complex bus
voltages based on the available set of measurements. Because phasor measurement
units (PMUs) are becoming more widely employed in transmission power systems, a
fast SE solver capable of exploiting PMUs' high sample rates is required. To
accomplish this, we present a method for training a model based on graph neural
networks (GNNs) to learn estimates from PMU voltage and current measurements,
which, once it is trained, has a linear computational complexity with respect
to the number of nodes in the power system. We propose an original GNN
implementation over the power system's factor graph to simplify the
incorporation of various types and numbers of measurements both on power system
buses and branches. Furthermore, we augment the factor graph to improve the
robustness of GNN predictions. Training and test examples were generated by
randomly sampling sets of power system measurements and annotated with the
exact solutions of linear SE with PMUs. The numerical results demonstrate that
the GNN model provides an accurate approximation of the SE solutions.
Additionally, errors caused by PMU malfunctions or the communication failures
that make the SE problem unobservable have a local effect and do not
deteriorate the results in the rest of the power system. |
2,677 | Neuro-Nav: A Library for Neurally-Plausible Reinforcement Learning | In this work we propose Neuro-Nav, an open-source library for neurally
plausible reinforcement learning (RL). RL is among the most common modeling
frameworks for studying decision making, learning, and navigation in biological
organisms. In utilizing RL, cognitive scientists often handcraft environments
and agents to meet the needs of their particular studies. On the other hand,
artificial intelligence researchers often struggle to find benchmarks for
neurally and biologically plausible representation and behavior (e.g., in
decision making or navigation). In order to streamline this process across both
fields with transparency and reproducibility, Neuro-Nav offers a set of
standardized environments and RL algorithms drawn from canonical behavioral and
neural studies in rodents and humans. We demonstrate that the toolkit
replicates relevant findings from a number of studies across both cognitive
science and RL literatures. We furthermore describe ways in which the library
can be extended with novel algorithms (including deep RL) and environments to
address future research needs of the field. |
2,678 | Revisiting Realistic Test-Time Training: Sequential Inference and Adaptation by Anchored Clustering | Deploying models on target domain data subject to distribution shift requires
adaptation. Test-time training (TTT) emerges as a solution to this adaptation
under a realistic scenario where access to full source domain data is not
available and instant inference on target domain is required. Despite many
efforts into TTT, there is a confusion over the experimental settings, thus
leading to unfair comparisons. In this work, we first revisit TTT assumptions
and categorize TTT protocols by two key factors. Among the multiple protocols,
we adopt a realistic sequential test-time training (sTTT) protocol, under which
we further develop a test-time anchored clustering (TTAC) approach to enable
stronger test-time feature learning. TTAC discovers clusters in both source and
target domain and match the target clusters to the source ones to improve
generalization. Pseudo label filtering and iterative updating are developed to
improve the effectiveness and efficiency of anchored clustering. We demonstrate
that under all TTT protocols TTAC consistently outperforms the state-of-the-art
methods on five TTT datasets. We hope this work will provide a fair
benchmarking of TTT methods and future research should be compared within
respective protocols. A demo code is available at
https://github.com/Gorilla-Lab-SCUT/TTAC. |
2,679 | FedNST: Federated Noisy Student Training for Automatic Speech Recognition | Federated Learning (FL) enables training state-of-the-art Automatic Speech
Recognition (ASR) models on user devices (clients) in distributed systems,
hence preventing transmission of raw user data to a central server. A key
challenge facing practical adoption of FL for ASR is obtaining ground-truth
labels on the clients. Existing approaches rely on clients to manually
transcribe their speech, which is impractical for obtaining large training
corpora. A promising alternative is using semi-/self-supervised learning
approaches to leverage unlabelled user data. To this end, we propose a new
Federated ASR method called FedNST for noisy student training of distributed
ASR models with private unlabelled user data. We explore various facets of
FedNST , such as training models with different proportions of unlabelled and
labelled data, and evaluate the proposed approach on 1173 simulated clients.
Evaluating FedNST on LibriSpeech, where 960 hours of speech data is split
equally into server (labelled) and client (unlabelled) data, showed a 22.5%
relative word error rate reduction (WERR) over a supervised baseline trained
only on server data. |
2,680 | FuSS: Fusing Superpixels for Improved Segmentation Consistency | In this work, we propose two different approaches to improve the semantic
consistency of Open Set Semantic Segmentation. First, we propose a method
called OpenGMM that extends the OpenPCS framework using a Gaussian Mixture of
Models to model the distribution of pixels for each class in a multimodal
manner. The second approach is a post-processing which uses superpixels to
enforce highly homogeneous regions to behave equally, rectifying erroneous
classified pixels within these regions, we also proposed a novel superpixel
method called FuSS. All tests were performed on ISPRS Vaihingen and Potsdam
datasets, and both methods were capable to improve quantitative and qualitative
results for both datasets. Besides that, the post-process with FuSS achieved
state-of-the-art results for both datasets. The official implementation is
available at: \url{https://github.com/iannunes/FuSS}. |
2,681 | Is a Modular Architecture Enough? | Inspired from human cognition, machine learning systems are gradually
revealing advantages of sparser and more modular architectures. Recent work
demonstrates that not only do some modular architectures generalize well, but
they also lead to better out-of-distribution generalization, scaling
properties, learning speed, and interpretability. A key intuition behind the
success of such systems is that the data generating system for most real-world
settings is considered to consist of sparsely interacting parts, and endowing
models with similar inductive biases will be helpful. However, the field has
been lacking in a rigorous quantitative assessment of such systems because
these real-world data distributions are complex and unknown. In this work, we
provide a thorough assessment of common modular architectures, through the lens
of simple and known modular data distributions. We highlight the benefits of
modularity and sparsity and reveal insights on the challenges faced while
optimizing modular systems. In doing so, we propose evaluation metrics that
highlight the benefits of modularity, the regimes in which these benefits are
substantial, as well as the sub-optimality of current end-to-end learned
modular systems as opposed to their claimed potential. |
2,682 | Specification-Guided Learning of Nash Equilibria with High Social Welfare | Reinforcement learning has been shown to be an effective strategy for
automatically training policies for challenging control problems. Focusing on
non-cooperative multi-agent systems, we propose a novel reinforcement learning
framework for training joint policies that form a Nash equilibrium. In our
approach, rather than providing low-level reward functions, the user provides
high-level specifications that encode the objective of each agent. Then, guided
by the structure of the specifications, our algorithm searches over policies to
identify one that provably forms an $\epsilon$-Nash equilibrium (with high
probability). Importantly, it prioritizes policies in a way that maximizes
social welfare across all agents. Our empirical evaluation demonstrates that
our algorithm computes equilibrium policies with high social welfare, whereas
state-of-the-art baselines either fail to compute Nash equilibria or compute
ones with comparatively lower social welfare. |
2,683 | Perturbation Learning Based Anomaly Detection | This paper presents a simple yet effective method for anomaly detection. The
main idea is to learn small perturbations to perturb normal data and learn a
classifier to classify the normal data and the perturbed data into two
different classes. The perturbator and classifier are jointly learned using
deep neural networks. Importantly, the perturbations should be as small as
possible but the classifier is still able to recognize the perturbed data from
unperturbed data. Therefore, the perturbed data are regarded as abnormal data
and the classifier provides a decision boundary between the normal data and
abnormal data, although the training data do not include any abnormal data.
Compared with the state-of-the-art of anomaly detection, our method does not
require any assumption about the shape (e.g. hypersphere) of the decision
boundary and has fewer hyper-parameters to determine. Empirical studies on
benchmark datasets verify the effectiveness and superiority of our method. |
2,684 | Stochastic Variance-Reduced Newton: Accelerating Finite-Sum Minimization with Large Batches | Stochastic variance reduction has proven effective at accelerating
first-order algorithms for solving convex finite-sum optimization tasks such as
empirical risk minimization. Incorporating additional second-order information
has proven helpful in further improving the performance of these first-order
methods. However, comparatively little is known about the benefits of using
variance reduction to accelerate popular stochastic second-order methods such
as Subsampled Newton. To address this, we propose Stochastic Variance-Reduced
Newton (SVRN), a finite-sum minimization algorithm which enjoys all the
benefits of second-order methods: simple unit step size, easily parallelizable
large-batch operations, and fast local convergence, while at the same time
taking advantage of variance reduction to achieve improved convergence rates
(per data pass) for smooth and strongly convex problems. We show that SVRN can
accelerate many stochastic second-order methods (such as Subsampled Newton) as
well as iterative least squares solvers (such as Iterative Hessian Sketch), and
it compares favorably to popular first-order methods with variance reduction. |
2,685 | Multi-Behavior Sequential Recommendation with Temporal Graph Transformer | Modeling time-evolving preferences of users with their sequential item
interactions, has attracted increasing attention in many online applications.
Hence, sequential recommender systems have been developed to learn the dynamic
user interests from the historical interactions for suggesting items. However,
the interaction pattern encoding functions in most existing sequential
recommender systems have focused on single type of user-item interactions. In
many real-life online platforms, user-item interactive behaviors are often
multi-typed (e.g., click, add-to-favorite, purchase) with complex cross-type
behavior inter-dependencies. Learning from informative representations of users
and items based on their multi-typed interaction data, is of great importance
to accurately characterize the time-evolving user preference. In this work, we
tackle the dynamic user-item relation learning with the awareness of
multi-behavior interactive patterns. Towards this end, we propose a new
Temporal Graph Transformer (TGT) recommendation framework to jointly capture
dynamic short-term and long-range user-item interactive patterns, by exploring
the evolving correlations across different types of behaviors. The new TGT
method endows the sequential recommendation architecture to distill dedicated
knowledge for type-specific behavior relational context and the implicit
behavior dependencies. Experiments on the real-world datasets indicate that our
method TGT consistently outperforms various state-of-the-art recommendation
methods. Our model implementation codes are available at
https://github.com/akaxlh/TGT. |
2,686 | Separable Self-attention for Mobile Vision Transformers | Mobile vision transformers (MobileViT) can achieve state-of-the-art
performance across several mobile vision tasks, including classification and
detection. Though these models have fewer parameters, they have high latency as
compared to convolutional neural network-based models. The main efficiency
bottleneck in MobileViT is the multi-headed self-attention (MHA) in
transformers, which requires $O(k^2)$ time complexity with respect to the
number of tokens (or patches) $k$. Moreover, MHA requires costly operations
(e.g., batch-wise matrix multiplication) for computing self-attention,
impacting latency on resource-constrained devices. This paper introduces a
separable self-attention method with linear complexity, i.e. $O(k)$. A simple
yet effective characteristic of the proposed method is that it uses
element-wise operations for computing self-attention, making it a good choice
for resource-constrained devices. The improved model, MobileViTv2, is
state-of-the-art on several mobile vision tasks, including ImageNet object
classification and MS-COCO object detection. With about three million
parameters, MobileViTv2 achieves a top-1 accuracy of 75.6% on the ImageNet
dataset, outperforming MobileViT by about 1% while running $3.2\times$ faster
on a mobile device.
Our source code is available at: \url{https://github.com/apple/ml-cvnets} |
2,687 | Risk-Sensitive Reinforcement Learning: Iterated CVaR and the Worst Path | In this paper, we study a novel episodic risk-sensitive Reinforcement
Learning (RL) problem, named Iterated CVaR RL, where the objective is to
maximize the tail of the reward-to-go at each step. Different from existing
risk-aware RL formulations, Iterated CVaR RL focuses on safety-at-all-time, by
enabling the agent to tightly control the risk of getting into catastrophic
situations at each stage, and is applicable to important risk-sensitive tasks
that demand strong safety guarantees throughout the decision process, such as
autonomous driving, clinical treatment planning and robotics. We investigate
Iterated CVaR RL with two performance metrics, i.e., Regret Minimization and
Best Policy Identification. For both metrics, we design efficient algorithms
ICVaR-RM and ICVaR-BPI, respectively, and provide matching upper and lower
bounds with respect to the number of episodes $K$. We also investigate an
interesting limiting case of Iterated CVaR RL, called Worst Path RL, where the
objective becomes to maximize the minimum possible cumulative reward, and
propose an efficient algorithm with constant upper and lower bounds. Finally,
the techniques we develop for bounding the change of CVaR due to the value
function shift and decomposing the regret via a distorted visitation
distribution are novel and can find applications in other risk-sensitive online
learning problems. |
2,688 | Enhancing Safe Exploration Using Safety State Augmentation | Safe exploration is a challenging and important problem in model-free
reinforcement learning (RL). Often the safety cost is sparse and unknown, which
unavoidably leads to constraint violations -- a phenomenon ideally to be
avoided in safety-critical applications. We tackle this problem by augmenting
the state-space with a safety state, which is nonnegative if and only if the
constraint is satisfied. The value of this state also serves as a distance
toward constraint violation, while its initial value indicates the available
safety budget. This idea allows us to derive policies for scheduling the safety
budget during training. We call our approach Simmer (Safe policy IMproveMEnt
for RL) to reflect the careful nature of these schedules. We apply this idea to
two safe RL problems: RL with constraints imposed on an average cost, and RL
with constraints imposed on a cost with probability one. Our experiments
suggest that simmering a safe algorithm can improve safety during training for
both settings. We further show that Simmer can stabilize training and improve
the performance of safe RL with average constraints. |
2,689 | Robust Adversarial Attacks Detection based on Explainable Deep Reinforcement Learning For UAV Guidance and Planning | The danger of adversarial attacks to unprotected Uncrewed Aerial Vehicle
(UAV) agents operating in public is growing. Adopting AI-based techniques and
more specifically Deep Learning (DL) approaches to control and guide these UAVs
can be beneficial in terms of performance but add more concerns regarding the
safety of those techniques and their vulnerability against adversarial attacks
causing the chances of collisions going up as the agent becomes confused. This
paper proposes an innovative approach based on the explainability of DL methods
to build an efficient detector that will protect these DL schemes and thus the
UAVs adopting them from potential attacks. The agent is adopting a Deep
Reinforcement Learning (DRL) scheme for guidance and planning. It is formed and
trained with a Deep Deterministic Policy Gradient (DDPG) with Prioritised
Experience Replay (PER) DRL scheme that utilises Artificial Potential Field
(APF) to improve training times and obstacle avoidance performance. The
adversarial attacks are generated by Fast Gradient Sign Method (FGSM) and Basic
Iterative Method (BIM) algorithms and reduced obstacle course completion rates
from 80\% to 35\%. A Realistic Synthetic environment for UAV explainable DRL
based planning and guidance including obstacles and adversarial attacks is
built. Two adversarial attack detectors are proposed. The first one adopts a
Convolutional Neural Network (CNN) architecture and achieves an accuracy in
detection of 80\%. The second detector is developed based on a Long Short Term
Memory (LSTM) network and achieves an accuracy of 91\% with much faster
computing times when compared to the CNN based detector. |
2,690 | Multi-learner risk reduction under endogenous participation dynamics | Prediction systems face exogenous and endogenous distribution shift -- the
world constantly changes, and the predictions the system makes change the
environment in which it operates. For example, a music recommender observes
exogeneous changes in the user distribution as different communities have
increased access to high speed internet. If users under the age of 18 enjoy
their recommendations, the proportion of the user base comprised of those under
18 may endogeneously increase. Most of the study of endogenous shifts has
focused on the single decision-maker setting, where there is one learner that
users either choose to use or not.
This paper studies participation dynamics between sub-populations and
possibly many learners. We study the behavior of systems with
\emph{risk-reducing} learners and sub-populations. A risk-reducing learner
updates their decision upon observing a mixture distribution of the
sub-populations $\mathcal{D}$ in such a way that it decreases the risk of the
learner on that mixture. A risk reducing sub-population updates its
apportionment amongst learners in a way which reduces its overall loss.
Previous work on the single learner case shows that myopic risk minimization
can result in high overall loss~\citep{perdomo2020performative,
miller2021outside} and representation disparity~\citep{hashimoto2018fairness,
zhang2019group}. Our work analyzes the outcomes of multiple myopic learners and
market forces, often leading to better global loss and less representation
disparity. |
2,691 | Robust Pareto Set Identification with Contaminated Bandit Feedback | We consider the Pareto set identification (PSI) problem in multi-objective
multi-armed bandits (MO-MAB) with contaminated reward observations. At each arm
pull, with some probability, the true reward samples are replaced with the
samples from an arbitrary contamination distribution chosen by the adversary.
We propose a median-based MO-MAB algorithm for robust PSI that abides by the
accuracy requirements set by the user via an accuracy parameter. We prove that
the sample complexity of this algorithm depends on the accuracy parameter
inverse squarely. We compare the proposed algorithm with a mean-based method
from MO-MAB literature on Gaussian reward distributions. Our numerical results
verify our theoretical expectations and show the necessity for robust algorithm
design in the adversarial setting. |
2,692 | Learning with Capsules: A Survey | Capsule networks were proposed as an alternative approach to Convolutional
Neural Networks (CNNs) for learning object-centric representations, which can
be leveraged for improved generalization and sample complexity. Unlike CNNs,
capsule networks are designed to explicitly model part-whole hierarchical
relationships by using groups of neurons to encode visual entities, and learn
the relationships between those entities. Promising early results achieved by
capsule networks have motivated the deep learning community to continue trying
to improve their performance and scalability across several application areas.
However, a major hurdle for capsule network research has been the lack of a
reliable point of reference for understanding their foundational ideas and
motivations. The aim of this survey is to provide a comprehensive overview of
the capsule network research landscape, which will serve as a valuable resource
for the community going forward. To that end, we start with an introduction to
the fundamental concepts and motivations behind capsule networks, such as
equivariant inference in computer vision. We then cover the technical advances
in the capsule routing mechanisms and the various formulations of capsule
networks, e.g. generative and geometric. Additionally, we provide a detailed
explanation of how capsule networks relate to the popular attention mechanism
in Transformers, and highlight non-trivial conceptual similarities between them
in the context of representation learning. Afterwards, we explore the extensive
applications of capsule networks in computer vision, video and motion, graph
representation learning, natural language processing, medical imaging and many
others. To conclude, we provide an in-depth discussion regarding the main
hurdles in capsule network research, and highlight promising research
directions for future work. |
2,693 | TransBO: Hyperparameter Optimization via Two-Phase Transfer Learning | With the extensive applications of machine learning models, automatic
hyperparameter optimization (HPO) has become increasingly important. Motivated
by the tuning behaviors of human experts, it is intuitive to leverage auxiliary
knowledge from past HPO tasks to accelerate the current HPO task. In this
paper, we propose TransBO, a novel two-phase transfer learning framework for
HPO, which can deal with the complementary nature among source tasks and
dynamics during knowledge aggregation issues simultaneously. This framework
extracts and aggregates source and target knowledge jointly and adaptively,
where the weights can be learned in a principled manner. The extensive
experiments, including static and dynamic transfer learning settings and neural
architecture search, demonstrate the superiority of TransBO over the
state-of-the-arts. |
2,694 | Port-Hamiltonian Neural Networks with State Dependent Ports | Hybrid machine learning based on Hamiltonian formulations has recently been
successfully demonstrated for simple mechanical systems. In this work, we
stress-test the method on both simple mass-spring systems and more complex and
realistic systems with several internal and external forces, including a system
with multiple connected tanks. We quantify performance under various conditions
and show that imposing different assumptions greatly affect the performance
during training presenting advantages and limitations of the method. We
demonstrate that port-Hamiltonian neural networks can be extended to larger
dimensions with state-dependent ports. We consider learning on systems with
known and unknown external forces and show how it can be used to detect
deviations in a system and still provide a valid model when the deviations are
removed. Finally, we propose a symmetric high-order integrator for improved
training on sparse and noisy data. |
2,695 | Robust Fine-Tuning of Deep Neural Networks with Hessian-based Generalization Guarantees | We consider transfer learning approaches that fine-tune a pretrained deep
neural network on a target task. We investigate generalization properties of
fine-tuning to understand the problem of overfitting, which often happens in
practice. Previous works have shown that constraining the distance from the
initialization of fine-tuning improves generalization. Using a PAC-Bayesian
analysis, we observe that besides distance from initialization, Hessians affect
generalization through the noise stability of deep neural networks against
noise injections. Motivated by the observation, we develop Hessian
distance-based generalization bounds for a wide range of fine-tuning methods.
Next, we investigate the robustness of fine-tuning with noisy labels. We design
an algorithm that incorporates consistent losses and distance-based
regularization for fine-tuning. Additionally, we prove a generalization error
bound of our algorithm under class conditional independent noise in the
training dataset labels. We perform a detailed empirical study of our algorithm
on various noisy environments and architectures. For example, on six image
classification tasks whose training labels are generated with programmatic
labeling, we show a 3.26% accuracy improvement over prior methods. Meanwhile,
the Hessian distance measure of the fine-tuned network using our algorithm
decreases by six times more than existing approaches. |
2,696 | A Regret-Variance Trade-Off in Online Learning | We consider prediction with expert advice for strongly convex and bounded
losses, and investigate trade-offs between regret and "variance" (i.e., squared
difference of learner's predictions and best expert predictions). With $K$
experts, the Exponentially Weighted Average (EWA) algorithm is known to achieve
$O(\log K)$ regret. We prove that a variant of EWA either achieves a negative
regret (i.e., the algorithm outperforms the best expert), or guarantees a
$O(\log K)$ bound on both variance and regret. Building on this result, we show
several examples of how variance of predictions can be exploited in learning.
In the online to batch analysis, we show that a large empirical variance allows
to stop the online to batch conversion early and outperform the risk of the
best predictor in the class. We also recover the optimal rate of model
selection aggregation when we do not consider early stopping. In online
prediction with corrupted losses, we show that the effect of corruption on the
regret can be compensated by a large variance. In online selective sampling, we
design an algorithm that samples less when the variance is large, while
guaranteeing the optimal regret bound in expectation. In online learning with
abstention, we use a similar term as the variance to derive the first
high-probability $O(\log K)$ regret bound in this setting. Finally, we extend
our results to the setting of online linear regression. |
2,697 | Crust Macrofracturing as the Evidence of the Last Deglaciation | Machine learning methods were applied to reconsider the results of several
passive seismic experiments in Finland. We created datasets from different
stages of the receiver function technique and processed them with one of basic
machine learning algorithms. All the results were obtained uniformly with the
$k$-nearest neighbors algorithm. The first result is the Moho depth map of the
region. Another result is the delineation of the near-surface low $S$-wave
velocity layer. There are three such areas in the Northern, Southern, and
central parts of the region. The low $S$-wave velocity in the Northern and
Southern areas can be linked to the geological structure. However, we attribute
the central low $S$-wave velocity area to a large number of water-saturated
cracks in the upper 1-5 km. Analysis of the structure of this area leads us to
the conclusion that macrofracturing was caused by the last deglaciation. |
2,698 | Mixed Graph Contrastive Network for Semi-Supervised Node Classification | Graph Neural Networks (GNNs) have achieved promising performance in
semi-supervised node classification in recent years. However, the problem of
insufficient supervision, together with representation collapse, largely limits
the performance of the GNNs in this field. To alleviate the collapse of node
representations in semi-supervised scenario, we propose a novel graph
contrastive learning method, termed Mixed Graph Contrastive Network (MGCN). In
our method, we improve the discriminative capability of the latent feature by
enlarging the margin of decision boundaries and improving the cross-view
consistency of the latent representation. Specifically, we first adopt an
interpolation-based strategy to conduct data augmentation in the latent space
and then force the prediction model to change linearly between samples. Second,
we enable the learned network to tell apart samples across two
interpolation-perturbed views through forcing the correlation matrix across
views to approximate an identity matrix. By combining the two settings, we
extract rich supervision information from both the abundant unlabeled nodes and
the rare yet valuable labeled nodes for discriminative representation learning.
Extensive experimental results on six datasets demonstrate the effectiveness
and the generality of MGCN compared to the existing state-of-the-art methods. |
2,699 | Policy Optimization for Markov Games: Unified Framework and Faster Convergence | This paper studies policy optimization algorithms for multi-agent
reinforcement learning. We begin by proposing an algorithm framework for
two-player zero-sum Markov Games in the full-information setting, where each
iteration consists of a policy update step at each state using a certain matrix
game algorithm, and a value update step with a certain learning rate. This
framework unifies many existing and new policy optimization algorithms. We show
that the state-wise average policy of this algorithm converges to an
approximate Nash equilibrium (NE) of the game, as long as the matrix game
algorithms achieve low weighted regret at each state, with respect to weights
determined by the speed of the value updates. Next, we show that this framework
instantiated with the Optimistic Follow-The-Regularized-Leader (OFTRL)
algorithm at each state (and smooth value updates) can find an
$\mathcal{\widetilde{O}}(T^{-5/6})$ approximate NE in $T$ iterations, which
improves over the current best $\mathcal{\widetilde{O}}(T^{-1/2})$ rate of
symmetric policy optimization type algorithms. We also extend this algorithm to
multi-player general-sum Markov Games and show an
$\mathcal{\widetilde{O}}(T^{-3/4})$ convergence rate to Coarse Correlated
Equilibria (CCE). Finally, we provide a numerical example to verify our theory
and investigate the importance of smooth value updates, and find that using
"eager" value updates instead (equivalent to the independent natural policy
gradient algorithm) may significantly slow down the convergence, even on a
simple game with $H=2$ layers. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.