Profiles2022-2024
Gal Vardi
TTI-Chicago and the Hebrew University of Jerusalem
8Papers
1Last author
9Unique co-authors
3First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers8
- An Agnostic View on the Cost of Overfitting in (Kernel) Ridge RegressionAuthor 3 of 4ICLR 2024 · 2 citations
- The Double-Edged Sword of Implicit Bias: Generalization vs. Robustness in ReLU NetworksAuthor 2 of 6NeurIPS 2023 · 25 citations
- Adversarial Examples Exist in Two-Layer ReLU Networks for Low Dimensional Linear SubspacesAuthor 3 of 3NeurIPS 2023 · 9 citations
- Implicit Bias in Leaky ReLU Networks Trained on High-Dimensional DataAuthor 3 of 8ICLR 2023 · 5 citations
- Reconstructing Training Data From Trained Neural NetworksAuthor 2 of 5NeurIPS 2022 · 196 citations
- On Margin Maximization in Linear and ReLU NetworksAuthor 1 of 3NeurIPS 2022 · 37 citations
- Gradient Methods Provably Converge to Non-Robust NetworksAuthor 1 of 3NeurIPS 2022 · 32 citations
- The Sample Complexity of One-Hidden-Layer Neural NetworksAuthor 1 of 3NeurIPS 2022 · 11 citations
