Good data augmentation is one of the key factors that lead to the empiri...
We propose an evolution strategies-based algorithm for estimating gradie...
Backdoor inversion, the process of finding a backdoor trigger inserted i...
Most work on the formal verification of neural networks has focused on
b...
Identifying statistical regularities in solutions to some tasks in multi...
Finetuning image-text models such as CLIP achieves state-of-the-art
accu...
Designing networks capable of attaining better performance with an incre...
This work concerns the development of deep networks that are certifiably...
Diffusion-based generative models are extremely effective in generating
...
Lipschitz constants are connected to many properties of neural networks,...
This paper introduces a unified framework for video action segmentation ...
Test-time adaptation (TTA) refers to adapting neural networks to distrib...
Recently, Miller et al. showed that a model's in-distribution (ID) accur...
Beyond achieving high performance across many vision tasks, multimodal m...
Empirical risk minimization (ERM) is known in practice to be non-robust ...
Many modern machine learning tasks require models with high tail perform...
We study the utility of the lexical translation model (IBM Model 1) for
...
Researchers have repeatedly shown that it is possible to craft adversari...
The rise of machine learning (ML) has created an explosion in the potent...
Recent work in adversarial attacks has developed provably robust methods...
Recent work has shown how to embed differentiable optimization problems ...
In this paper, we demonstrate a physical adversarial patch attack agains...
Integrating logical reasoning within deep learning architectures has bee...