Sensing and communication technologies have enhanced learning-based deci...
Decisions made by machine learning models may have lasting impacts over ...
In reinforcement learning (RL), a reward function is often assumed at th...
CLIP, as a foundational vision language model, is widely used in zero-sh...
Robust reinforcement learning (RL) seeks to train policies that can perf...
Despite recent progress in reinforcement learning (RL) from raw pixel da...
Shift equivariance is a fundamental principle that governs how we percei...
Locality-sensitive hashing (LSH) based frameworks have been used efficie...
Most existing works consider direct perturbations of victim's state/acti...
Given the growing concerns about fairness in machine learning and the
im...
Our work focuses on the challenge of detecting outputs generated by Larg...
The robustness of a deep classifier can be characterized by its margins:...
Probabilistic dynamics model ensemble is widely used in existing model-b...
Directed Exploration is a crucial challenge in reinforcement learning (R...
Self-supervised pretraining has been extensively studied in language and...
Large-scale non-convex optimization problems are expensive to solve due ...
Data augmentation is a critical contributing factor to the success of de...
The decentralized Federated Learning (FL) setting avoids the role of a
p...
Recent studies reveal that a well-trained deep reinforcement learning (R...
Specializing Directed Acyclic Graph Federated Learning(SDAGFL) is a new
...
Standard diffusion models involve an image transform – adding Gaussian n...
Model-based reinforcement learning (RL) achieves higher sample efficienc...
The increasing reliance on ML models in high-stakes tasks has raised a m...
In federated learning (FL), the objective of collaboratively learning a
...
Communication is important in many multi-agent reinforcement learning (M...
In this work, we propose a novel Kernelized Stein
Discrepancy-based Post...
Machine learning systems perform well on pattern matching tasks, but the...
In many reinforcement learning (RL) applications, the observation space ...
Machine learning models that are developed to be invariant under certain...
Most state-of-the-art Graph Neural Networks (GNNs) can be defined as a f...
A popular application of federated learning is using many clients to tra...
The power method is a classical algorithm with broad applications in mac...
We describe new datasets for studying generalization from easy to hard
e...
Conventional saliency maps highlight input features to which neural netw...
Randomized Smoothing (RS), being one of few provable defenses, has been
...
Enforcing orthogonality in neural networks is an antidote for gradient
v...
Evaluating the worst-case performance of a reinforcement learning (RL) a...
Deep neural networks are powerful machines for visual pattern recognitio...
For deep learning practitioners, hyperparameter tuning for optimizing mo...
Data poisoning and backdoor attacks manipulate training data to induce
s...
Adversarial Training is proved to be an efficient method to defend again...
Poisoning attacks, although have been studied extensively in supervised
...
Adaptive gradient methods such as RMSProp and Adam use exponential movin...
Deep learning models extract, before a final classification layer, featu...
Convex relaxations are effective for training and certifying neural netw...
Higher-order Recurrent Neural Networks (RNNs) are effective for long-ter...
Transferring knowledge among various environments is important to effici...
Deep neural networks generalize well on unseen data though the number of...
Model-based reinforcement learning algorithms make decisions by building...
Bayesian learning of model parameters in neural networks is important in...