Do Perceptually Aligned Gradients Imply Robustness?
Jul 24, 2023
Sprecher:innen
Bahjat Kawar
Sprecher:in · 0 Follower:innen
Michael Elad
Sprecher:in · 0 Follower:innen
Über
Adversarially robust classifiers possess a trait that non-robust models do not - Perceptually Aligned Gradients (PAG). Their gradients with respect to the input align well with human perception. Several works have identified PAG as a byproduct of robust training, but none have considered it as a standalone phenomenon nor studied its own implications. In this work, we focus on this trait and test whether Perceptually Aligned Gradients imply Robustness. To this end, we develop a novel objective to directly promote PAG in training classifiers and examine whether models with such gradients are more robust to adversarial attacks. Extensive experiments on multiple datasets and architectures validate that models with aligned gradients exhibit significant robustness, exposing the surprising bidirectional connection between PAG and robustness.Lastly, we show that better gradient alignment leads to increased robustness and harness this observation to boost the robustness of existing adversarial training techniques.Adversarially robust classifiers possess a trait that non-robust models do not - Perceptually Aligned Gradients (PAG). Their gradients with respect to the input align well with human perception. Several works have identified PAG as a byproduct of robust training, but none have considered it as a standalone phenomenon nor studied its own implications. In this work, we focus on this trait and test whether Perceptually Aligned Gradients imply Robustness. To this end, we develop a novel objective to…
Gefällt euch das Format? Vertraut auf SlidesLive, um euer nächstes Event festzuhalten!
Professionelle Aufzeichnung und Livestreaming – weltweit.
Freigeben
Empfohlene Videos
Präsentationen, deren Thema, Kategorie oder Sprecher:in ähnlich sind
Geometric ML for molecules
Später ansehen
Recent Advances in the Generalization Theory of Neural Networks *
Später ansehen
Tengyu Ma, …
Conditional Diffusion Replay for Continual Learning in Medical Settings
Später ansehen
Yewon Byun, …
In-Context Learning for Bayesian Optimization
Später ansehen
Returning The Favour: When Regression Benefits From Probabilistic Causal Knowledge
Später ansehen
