Unrolling sgd: Understanding factors influencing machine unlearning A Thudi, G Deza, V Chandrasekaran, N Papernot 7th IEEE European Symposium on Security and Privacy, 2021 | 119 | 2021 |
On the Necessity of Auditable Algorithmic Definitions for Machine Unlearning A Thudi, H Jia, I Shumailov, N Papernot 31st USENIX Security Symposium, 2021 | 107 | 2021 |
Proof-of-learning: Definitions and practice H Jia, M Yaghini, CA Choquette-Choo, N Dullerud, A Thudi, ... 2021 IEEE Symposium on Security and Privacy (SP), 1039-1056, 2021 | 94 | 2021 |
Selective classification via neural network training dynamics S Rabanser, A Thudi, K Hamidieh, A Dziedzic, N Papernot arXiv preprint arXiv:2205.13532, 2022 | 20 | 2022 |
SoK: Machine learning governance V Chandrasekaran, H Jia, A Thudi, A Travers, M Yaghini, N Papernot arXiv preprint arXiv:2109.10870, 2021 | 20 | 2021 |
From Differential Privacy to Bounds on Membership Inference: Less can be More A Thudi, I Shumailov, F Boenisch, N Papernot Transactions on Machine Learning Research, 0 | 18* | |
Proof-of-learning is currently more broken than you think C Fang, H Jia, A Thudi, M Yaghini, CA Choquette-Choo, N Dullerud, ... 2023 IEEE 8th European Symposium on Security and Privacy (EuroS&P), 797-816, 2023 | 17* | 2023 |
Unlearnable algorithms for in-context learning A Muresanu, A Thudi, MR Zhang, N Papernot arXiv preprint arXiv:2402.00751, 2024 | 4 | 2024 |
Training Private Models That Know What They Don't Know S Rabanser, A Thudi, A Thakurta, K Dvijotham, N Papernot 37th Conference on Neural Information Processing Systems, 2023 | 3 | 2023 |
Better Sparsifiers for Directed Eulerian Graphs S Sachdeva, A Thudi, Y Zhao Proceedings of the 51st EATCS International Colloquium on Automata …, 2023 | 2 | 2023 |
Finding Optimally Robust Data Mixtures via Concave Maximization A Thudi, CJ Maddison arXiv preprint arXiv:2406.01477, 2024 | | 2024 |
Gradients Look Alike: Sensitivity is Often Overestimated in DP-SGD A Thudi, H Jia, C Meehan, I Shumailov, N Papernot Proceedings of the 33rd USENIX Security Symposium, 2023 | | 2023 |