The ability to generate privacy-preserving synthetic versions of sensiti...
Adversarial training is widely used to make classifiers robust to a spec...
Automated content filtering and moderation is an important tool that all...
We introduce the Lossy Implicit Network Activation Coding (LINAC) defenc...
While adversarial training is generally used as a defense mechanism, rec...
Automatically discovering failures in vision models under real-world set...
Recent work introduced the epinet as a new approach to uncertainty model...
Adaptive defenses that use test-time optimization promise to improve
rob...
Programming is a powerful and ubiquitous problem-solving tool. Developin...
Adversarial training suffers from robust overfitting, a phenomenon where...
Robustness to distribution shifts is critical for deploying machine lear...
Recent work argues that robust training requires substantially larger
da...
Collecting annotations from human raters often results in a trade-off be...
We study the adversarial robustness of information bottleneck models for...
Modern neural networks excel at image classification, yet they remain
vu...
Adversarial training suffers from robust overfitting, a phenomenon where...
We propose a general framework for verifying input-output specifications...
Does a Variational AutoEncoder (VAE) consistently encode typical samples...
Adversarial training and its variants have become de facto standards for...
Reinforcement learning (RL) has proven its worth in a series of artifici...
Recent research has made the surprising finding that state-of-the-art de...
In this paper we propose to augment a modern neural-network architecture...
Adversarial testing methods based on Projected Gradient Descent (PGD) ar...
Neural networks are part of many contemporary NLP systems, yet their
emp...
Adversarial training is an effective methodology for training deep neura...
Prior work on neural network verification has focused on specifications ...
Recent works have shown that it is possible to train models that are
ver...
Optimizing for long term value is desirable in many practical applicatio...
This paper proposes a new algorithmic framework,predictor-verifier
train...
This paper addresses the problem of formally verifying desirable propert...
The slate recommendation problem aims to find the "optimal" ordering of ...