Jailbroken: How does llm safety training fail? A Wei, N Haghtalab, J Steinhardt Advances in Neural Information Processing Systems 36, 2024 | 322 | 2024 |
Commitment Without Regrets: Online Learning in Stackelberg Security Games MF Balcan, A Blum, N Haghtalab, AD Procaccia | 137 | 2015 |
Learning optimal commitment to overcome insecurity A Blum, N Haghtalab, AD Procaccia Advances in Neural Information Processing Systems 27, 2014 | 111 | 2014 |
Efficient learning of linear separators under bounded noise P Awasthi, MF Balcan, N Haghtalab, R Urner Conference on Learning Theory, 167-190, 2015 | 101 | 2015 |
Learning and 1-bit compressed sensing under asymmetric noise P Awasthi, MF Balcan, N Haghtalab, H Zhang Conference on Learning Theory, 152-192, 2016 | 98 | 2016 |
The disparate equilibria of algorithmic decision making when individuals invest rationally LT Liu, A Wilson, N Haghtalab, AT Kalai, C Borgs, J Chayes Proceedings of the 2020 Conference on Fairness, Accountability, and …, 2020 | 86 | 2020 |
Ignorance is almost bliss: Near-optimal stochastic matching with few queries A Blum, JP Dickerson, N Haghtalab, AD Procaccia, T Sandholm, ... Proceedings of the Sixteenth ACM Conference on Economics and Computation …, 2015 | 82 | 2015 |
Oracle-efficient online learning and auction design M Dudík, N Haghtalab, H Luo, RE Schapire, V Syrgkanis, JW Vaughan Journal of the ACM (JACM) 67 (5), 1-57, 2020 | 77 | 2020 |
Maximizing welfare with incentive-aware evaluation mechanisms N Haghtalab, N Immorlica, B Lucier, JZ Wang arXiv preprint arXiv:2011.01956, 2020 | 69 | 2020 |
Collaborative PAC learning A Blum, N Haghtalab, AD Procaccia, M Qiao Advances in Neural Information Processing Systems 30, 2017 | 61 | 2017 |
The provable virtue of laziness in motion planning N Haghtalab, S Mackenzie, A Procaccia, O Salzman, S Srinivasa Proceedings of the International Conference on Automated Planning and …, 2018 | 56 | 2018 |
Smoothed analysis of online and differentially private learning N Haghtalab, T Roughgarden, A Shetty Advances in Neural Information Processing Systems 33, 9203-9215, 2020 | 50 | 2020 |
Smoothed analysis with adaptive adversaries N Haghtalab, T Roughgarden, A Shetty 2021 IEEE 62nd Annual Symposium on Foundations of Computer Science (FOCS …, 2022 | 45 | 2022 |
Online learning with a hint O Dekel, N Haghtalab, P Jaillet Advances in Neural Information Processing Systems 30, 2017 | 45 | 2017 |
Three strategies to success: Learning adversary models in security games N Haghtalab, F Fang, TH Nguyen, A Sinha, AD Procaccia, M Tambe | 45 | 2016 |
One for one, or all for all: Equilibria and optimality of collaboration in federated learning A Blum, N Haghtalab, RL Phillips, H Shao International Conference on Machine Learning, 1005-1014, 2021 | 41 | 2021 |
Structured robust submodular maximization: Offline and online algorithms N Anari, N Haghtalab, S Naor, S Pokutta, M Singh, A Torrico The 22nd International Conference on Artificial Intelligence and Statistics …, 2019 | 39 | 2019 |
Clustering in the Presence of Background Noise S Ben-David, N Haghtalab International Conference in Machine Learning (ICML 2014), 2014 | 35 | 2014 |
On-demand sampling: Learning optimally from multiple distributions N Haghtalab, M Jordan, E Zhao Advances in Neural Information Processing Systems 35, 406-419, 2022 | 32 | 2022 |
Lazy Defenders Are Almost Optimal Against Diligent Attackers A Blum, N Haghtalab, AD Procaccia 28th AAAI Conference on Artificial Intelligence, 2014 | 30 | 2014 |