 Geometric Inductive Biases of Deep Networks: The Role of Data and ArchitectureInternational Conference on Learning Representations (ICLR), 2024 |
 From Tempered to Benign Overfitting in ReLU Neural NetworksNeural Information Processing Systems (NeurIPS), 2023 |
 Task Arithmetic in the Tangent Space: Improved Editing of Pre-Trained
ModelsNeural Information Processing Systems (NeurIPS), 2023 |
 Boosting Adversarial Robustness From The Perspective of Effective Margin
RegularizationBritish Machine Vision Conference (BMVC), 2022 |
 Integral Probability Metrics PAC-Bayes BoundsNeural Information Processing Systems (NeurIPS), 2022 |
 Generalization Through The Lens Of Leave-One-Out ErrorInternational Conference on Learning Representations (ICLR), 2022 |
 What can linearized neural networks actually say about generalization?Neural Information Processing Systems (NeurIPS), 2021 |