Obserwuj
Ayush Sekhari
Ayush Sekhari
Postdoctoral Associate, MIT
Zweryfikowany adres z mit.edu - Strona główna
Tytuł
Cytowane przez
Cytowane przez
Rok
Remember what you want to forget: Algorithms for machine unlearning
A Sekhari, J Acharya, G Kamath, AT Suresh
Advances in Neural Information Processing Systems 34, 18075-18086, 2021
3042021
Hybrid rl: Using both offline and online data can make rl efficient
Y Song, Y Zhou, A Sekhari, JA Bagnell, A Krishnamurthy, W Sun
arXiv preprint arXiv:2210.06718, 2022
932022
Uniform convergence of gradients for non-convex learning and optimization
DJ Foster, A Sekhari, K Sridharan
Advances in neural information processing systems 31, 2018
902018
Guarantees for epsilon-greedy reinforcement learning with function approximation
C Dann, Y Mansour, M Mohri, A Sekhari, K Sridharan
International conference on machine learning, 4666-4689, 2022
722022
Second-order information in non-convex stochastic optimization: Power and limitations
Y Arjevani, Y Carmon, JC Duchi, DJ Foster, A Sekhari, K Sridharan
Conference on Learning Theory, 242-299, 2020
672020
The complexity of making the gradient small in stochastic convex optimization
DJ Foster, A Sekhari, O Shamir, N Srebro, K Sridharan, B Woodworth
Conference on Learning Theory, 1319-1345, 2019
662019
Hidden poison: Machine unlearning enables camouflaged poisoning attacks
JZ Di, J Douglas, J Acharya, G Kamath, A Sekhari
NeurIPS ML Safety Workshop, 2022
462022
Sgd: The role of implicit regularization, batch-size and multiple-epochs
A Sekhari, K Sridharan, S Kale
Advances In Neural Information Processing Systems 34, 27422-27433, 2021
442021
Provably efficient reinforcement learning in partially observable dynamical systems
M Uehara, A Sekhari, JD Lee, N Kallus, W Sun
Advances in Neural Information Processing Systems 35, 578-592, 2022
402022
On the complexity of adversarial decision making
DJ Foster, A Rakhlin, A Sekhari, K Sridharan
Advances in Neural Information Processing Systems 35, 35404-35417, 2022
382022
Neural active learning with performance guarantees
Z Wang, P Awasthi, C Dann, A Sekhari, C Gentile
Advances in Neural Information Processing Systems 34, 7510-7521, 2021
262021
Contextual bandits and imitation learning with preference-based active queries
A Sekhari, K Sridharan, W Sun, R Wu
Advances in Neural Information Processing Systems 36, 11261-11295, 2023
202023
Harnessing density ratios for online reinforcement learning
P Amortila, DJ Foster, N Jiang, A Sekhari, T Xie
arXiv preprint arXiv:2401.09681, 2024
152024
Agnostic reinforcement learning with low-rank MDPs and rich observations
A Sekhari, C Dann, M Mohri, Y Mansour, K Sridharan
Advances in Neural Information Processing Systems 34, 19033-19045, 2021
152021
Machine unlearning fails to remove data poisoning attacks
M Pawelczyk, JZ Di, Y Lu, G Kamath, A Sekhari, S Neel
arXiv preprint arXiv:2406.17216, 2024
142024
Computationally efficient pac rl in pomdps with latent determinism and conditional embeddings
M Uehara, A Sekhari, JD Lee, N Kallus, W Sun
International Conference on Machine Learning, 34615-34641, 2023
142023
A note on model-free reinforcement learning with the decision-estimation coefficient
DJ Foster, N Golowich, J Qian, A Rakhlin, A Sekhari
arXiv preprint arXiv:2211.14250, 2022
142022
Model-free reinforcement learning with the decision-estimation coefficient
DJ Foster, N Golowich, J Qian, A Rakhlin, A Sekhari
Advances in Neural Information Processing Systems 36, 20080-20117, 2023
132023
Reinforcement learning with feedback graphs
C Dann, Y Mansour, M Mohri, A Sekhari, K Sridharan
Advances in Neural Information Processing Systems 33, 16868-16878, 2020
132020
From Gradient Flow on Population Loss to Learning with Stochastic Gradient Descent
S Kale, JD Lee, C De Sa, A Sekhari, K Sridharan
arXiv preprint arXiv:2210.06705, 2022
12*2022
Nie można teraz wykonać tej operacji. Spróbuj ponownie później.
Prace 1–20