The problem of Byzantine resilience in distributed machine learning, aka Byzantine machine
learning, consists of designing distributed algorithms that can train an accurate model …

Despite the rapid increase of data available to train machine-learning algorithms in many
domains, several applications suffer from a paucity of representative and diverse data. The …

Adversarial purification refers to a class of defense methods that remove adversarial
perturbations using a generative model. These methods do not make assumptions on the …

While adversarial training is considered as a standard defense method against adversarial
attacks for image classifiers, adversarial purification, which purifies attacked images into …

Numerous deep learning algorithms have been inspired by and understood via the notion of
information bottleneck, where unnecessary information is (often implicitly) minimized while …

Adversarial Training (AT) has become arguably the state-of-the-art algorithm for extracting
robust features. However, researchers recently notice that AT suffers from severe robust …

Abstract No-Reference Video Quality Assessment (NR-VQA) plays an essential role in
improving the viewing experience of end-users. Driven by deep learning, recent NR-VQA …

The Lipschitz constant of neural networks has been established as a key quantity to enforce
the robustness to adversarial examples. In this paper, we tackle the problem of building $1 …

A variety of fairness constraints have been proposed in the literature to mitigate group-level
statistical bias. Their impacts have been largely evaluated for different groups of populations …

The accuracy of modern machine learning algorithms deteriorates severely on adversarially
manipulated test data. Optimal adversarial risk quantifies the best error rate of any classifier …