Sotabase

Career

· Professor, ETH Zurich2025–
· PhD Student, UC Berkeley

Publications (49)

Adversarial Training Can Hurt Generalization
arXiv.org · 2019
253
cited
Understanding and Mitigating the Tradeoff Between Robustness and Accuracy
International Conference on Machine Learning · 2020
246
cited
Regularized Learning for Domain Adaptation under Label Shifts
International Conference on Learning Representations · 2019
219
cited
Early Stopping for Kernel Boosting Algorithms: A General Analysis With Localized Complexities
IEEE Transactions on Information Theory · 2017
81
cited
Online control of the false discovery rate with decaying memory
Neural Information Processing Systems · 2017
71
cited
A framework for Multi-A(rmed)/B(andit) Testing with Online FDR Control
Neural Information Processing Systems · 2017
66
cited
Statistical and computational guarantees for the Baum-Welch algorithm
Allerton Conference on Communication, Control, and Computing · 2015
47
cited
Invariance-inducing regularization using worst-case transformations suffices to boost accuracy and spatial robustness
Neural Information Processing Systems · 2019
46
cited
Phaseless Signal Recovery in Infinite Dimensional Spaces Using Structured Modulations
arXiv.org · 2013
43
cited
How rotational invariance of common kernels prevents generalization in high dimensions
International Conference on Machine Learning · 2021
31
cited
How unfair is private learning ?
Conference on Uncertainty in Artificial Intelligence · 2022
26
cited
Fast rates for noisy interpolation require rethinking the effects of inductive bias
International Conference on Machine Learning · 2022
25
cited
Why adversarial training can hurt robust accuracy
International Conference on Learning Representations · 2022
22
cited
Phase retrieval from low rate samples
arXiv.org · 2013
20
cited
Phase Retrieval via Structured Modulations in Paley-Wiener Spaces
arXiv.org · 2013
20
cited
Self-supervised Reinforcement Learning with Independently Controllable Subgoals
Conference on Robot Learning · 2021
20
cited
Tight bounds for minimum l1-norm interpolation of noisy data
arXiv.org · 2021
20
cited
Interpolation can hurt robust generalization even when there is no noise
Neural Information Processing Systems · 2021
16
cited
PILLAR: How to make semi-private learning more effective
2024 IEEE Conference on Secure and Trustworthy Machine Learning (SaTML) · 2023
14
cited
Strong inductive biases provably prevent harmless interpolation
International Conference on Learning Representations · 2023
10
cited
Sotabase
Fanny Yang | Researcher Profile | Sotabase | Sotabase