Picture for Gal Vardi

Gal Vardi

Most Neural Networks Are Almost Learnable

May 25, 2023
Amit Daniely, Nathan Srebro, Gal Vardi

Access Paper or Ask Questions

Reconstructing Training Data from Multiclass Neural Networks

May 05, 2023
Gon Buzaglo, Niv Haim, Gilad Yehudai, Gal Vardi, Michal Irani

Access Paper or Ask Questions

Benign Overfitting in Linear Classifiers and Leaky ReLU Networks from KKT Conditions for Margin Maximization

Mar 02, 2023
Spencer Frei, Gal Vardi, Peter L. Bartlett, Nathan Srebro

* 53 pages

Access Paper or Ask Questions

The Double-Edged Sword of Implicit Bias: Generalization vs. Robustness in ReLU Networks

Mar 02, 2023
Spencer Frei, Gal Vardi, Peter L. Bartlett, Nathan Srebro

* 41 pages

Access Paper or Ask Questions

Efficiently Learning Neural Networks: What Assumptions May Suffice?

Feb 15, 2023
Amit Daniely, Nathan Srebro, Gal Vardi

* arXiv admin note: text overlap with arXiv:2101.08303

Access Paper or Ask Questions

On the Implicit Bias in Deep-Learning Algorithms

Aug 26, 2022
Gal Vardi

Figure 1 for On the Implicit Bias in Deep-Learning Algorithms

Figure 2 for On the Implicit Bias in Deep-Learning Algorithms

Access Paper or Ask Questions

On the Effective Number of Linear Regions in Shallow Univariate ReLU Networks: Convergence Guarantees and Implicit Bias

May 18, 2022
Itay Safran, Gal Vardi, Jason D. Lee

Figure 1 for On the Effective Number of Linear Regions in Shallow Univariate ReLU Networks: Convergence Guarantees and Implicit Bias

Figure 2 for On the Effective Number of Linear Regions in Shallow Univariate ReLU Networks: Convergence Guarantees and Implicit Bias

Access Paper or Ask Questions