Timezone: »
Given the apparent difficulty of learning models that are robust to adversarial perturbations, we propose tackling the simpler problem of developing adversarially robust features. Specifically, given a dataset and metric of interest, the goal is to return a function (or multiple functions) that 1) is robust to adversarial perturbations, and 2) has significant variation across the datapoints. We establish strong connections between adversarially robust features and a natural spectral property of the geometry of the dataset and metric of interest. This connection can be leveraged to provide both robust features, and a lower bound on the robustness of any function that has significant variance across the dataset. Finally, we provide empirical evidence that the adversarially robust features given by this spectral approach can be fruitfully leveraged to learn a robust (and accurate) model.
Author Information
Shivam Garg (Stanford University)
Vatsal Sharan (Stanford University)
Brian Zhang (Stanford University)
Gregory Valiant (Stanford University)
Related Events (a corresponding poster, oral, or spotlight)
-
2018 Poster: A Spectral View of Adversarially Robust Features »
Tue Dec 4th through Wed the 5th Room Room 517 AB
More from the Same Authors
-
2020 Poster: Small Nash Equilibrium Certificates in Very Large Games »
Brian Zhang · Tuomas Sandholm -
2019 Poster: Making AI Forget You: Data Deletion in Machine Learning »
Antonio Ginart · Melody Guan · Gregory Valiant · James Zou -
2019 Spotlight: Making AI Forget You: Data Deletion in Machine Learning »
Antonio Ginart · Melody Guan · Gregory Valiant · James Zou -
2019 Poster: PIDForest: Anomaly Detection via Partial Identification »
Parikshit Gopalan · Vatsal Sharan · Udi Wieder -
2019 Spotlight: PIDForest: Anomaly Detection via Partial Identification »
Parikshit Gopalan · Vatsal Sharan · Udi Wieder -
2018 Poster: Efficient Anomaly Detection via Matrix Sketching »
Vatsal Sharan · Parikshit Gopalan · Udi Wieder -
2017 Poster: Learning Overcomplete HMMs »
Vatsal Sharan · Sham Kakade · Percy Liang · Gregory Valiant -
2013 Poster: Estimating the Unseen: Improved Estimators for Entropy and other Properties »
Paul Valiant · Gregory Valiant