| Toward deeper understanding of neural networks: The power of initialization and a dual view on expressivity A Daniely, R Frostig, Y Singer Advances in neural information processing systems 29, 2016 | 390 | 2016 |
| SGD learns the conjugate kernel class of the network A Daniely Advances in neural information processing systems 30, 2017 | 196 | 2017 |
| Strongly adaptive online learning A Daniely, A Gonen, S Shalev-Shwartz International Conference on Machine Learning, 1405-1411, 2015 | 193 | 2015 |
| Complexity theoretic limitations on learning halfspaces A Daniely Proceedings of the forty-eighth annual ACM symposium on Theory of Computing …, 2016 | 150 | 2016 |
| Complexity theoretic limitations on learning dnf’s A Daniely, S Shalev-Shwartz Conference on Learning Theory, 815-830, 2016 | 120 | 2016 |
| From average case complexity to improper learning complexity A Daniely, N Linial, S Shalev-Shwartz Proceedings of the forty-sixth annual ACM symposium on Theory of computing …, 2014 | 118 | 2014 |
| Depth separation for neural networks A Daniely Conference on Learning Theory, 690-696, 2017 | 99 | 2017 |
| Optimal learners for multiclass problems A Daniely, S Shalev-Shwartz Conference on Learning Theory, 287-316, 2014 | 99 | 2014 |
| Learning parities with neural networks A Daniely, E Malach Advances in Neural Information Processing Systems 33, 20356-20365, 2020 | 94 | 2020 |
| Multiclass learnability and the ERM principle. A Daniely, S Sabato, S Ben-David, S Shalev-Shwartz J. Mach. Learn. Res. 16 (1), 2377-2404, 2015 | 93 | 2015 |
| Multiclass learnability and the erm principle A Daniely, S Sabato, S Ben-David, S Shalev-Shwartz Proceedings of the 24th Annual Conference on Learning Theory, 207-232, 2011 | 88 | 2011 |
| The implicit bias of depth: How incremental learning drives generalization D Gissin, S Shalev-Shwartz, A Daniely arXiv preprint arXiv:1909.12051, 2019 | 78 | 2019 |
| Multiclass learning approaches: A theoretical comparison with implications A Daniely, S Sabato, S Shwartz Advances in Neural Information Processing Systems 25, 2012 | 58 | 2012 |
| A PTAS for agnostically learning halfspaces A Daniely Conference on Learning Theory, 484-502, 2015 | 55 | 2015 |
| Learning economic parameters from revealed preferences MF Balcan, A Daniely, R Mehta, R Urner, VV Vazirani Web and Internet Economics: 10th International Conference, WINE 2014 …, 2014 | 53 | 2014 |
| Clustering is difficult only when it does not matter A Daniely, N Linial, M Saks arXiv preprint arXiv:1205.4891, 2012 | 49 | 2012 |
| On the practically interesting instances of MAXCUT Y Bilu, A Daniely, N Linial, M Saks arXiv preprint arXiv:1205.4893, 2012 | 47 | 2012 |
| More data speeds up training time in learning halfspaces over sparse vectors A Daniely, N Linial, S Shalev-Shwartz Advances in Neural Information Processing Systems 26, 2013 | 43 | 2013 |
| Most ReLU Networks Suffer from Adversarial Perturbations A Daniely, H Shacham Advances in Neural Information Processing Systems 33, 6629-6636, 2020 | 34 | 2020 |
| Neural networks learning and memorization with (almost) no over-parameterization A Daniely Advances in Neural Information Processing Systems 33, 9007-9016, 2020 | 34 | 2020 |
Shai Shalev-ShwartzThe Hebrew UniversityVerified email at cs.huji.ac.il
