Følg
Saeed Mahloujifar
Saeed Mahloujifar
FAIR, Meta
Verifisert e-postadresse på meta.com - Startside
Tittel
Sitert av
Sitert av
År
Improving adversarial robustness using proxy distributions
V Sehwag, S Mahloujifar, T Handina, S Dai, C Xiang, M Chiang, P Mittal
arXiv preprint arXiv:2104.09425 1, 2021
208*2021
The curse of concentration in robust learning: Evasion and poisoning attacks from concentration of measure
S Mahloujifar, DI Diochnos, M Mahmoody
Proceedings of the AAAI Conference on Artificial Intelligence 33, 4536-4543, 2019
1712019
Sparsefed: Mitigating model poisoning attacks in federated learning with sparsification
A Panda, S Mahloujifar, AN Bhagoji, S Chakraborty, P Mittal
International Conference on Artificial Intelligence and Statistics, 7587-7624, 2022
1112022
Revisiting the assumption of latent separability for backdoor defenses
X Qi, T Xie, Y Li, S Mahloujifar, P Mittal
The eleventh international conference on learning representations, 2023
1102023
Property inference from poisoning
S Mahloujifar, E Ghosh, M Chase
2022 IEEE Symposium on Security and Privacy (SP), 1120-1137, 2022
106*2022
Mitigating membership inference attacks by {Self-Distillation} through a novel ensemble architecture
X Tang, S Mahloujifar, L Song, V Shejwalkar, M Nasr, A Houmansadr, ...
31st USENIX security symposium (USENIX security 22), 1433-1450, 2022
1042022
Is private learning possible with instance encoding?
N Carlini, S Deng, S Garg, S Jha, S Mahloujifar, M Mahmoody, A Thakurta, ...
2021 IEEE Symposium on Security and Privacy (SP), 410-427, 2021
952021
Adversarial risk and robustness: General definitions and implications for the uniform distribution
D Diochnos, S Mahloujifar, M Mahmoody
Advances in Neural Information Processing Systems 31, 2018
942018
{PatchCleanser}: Certifiably robust defense against adversarial patches for any image classifier
C Xiang, S Mahloujifar, P Mittal
31st USENIX security symposium (USENIX Security 22), 2065-2082, 2022
882022
Universal Multi-Party Poisoning Attacks
S Mahloujifar, M Mahmoody, A Mohammed
International Conference on Machine Learning, 4274-4283, 2019
88*2019
Model-targeted poisoning attacks with provable convergence
F Suya, S Mahloujifar, A Suri, D Evans, Y Tian
International Conference on Machine Learning, 10000-10010, 2021
552021
Towards a proactive {ML} approach for detecting backdoor poison samples
X Qi, T Xie, JT Wang, T Wu, S Mahloujifar, P Mittal
32nd USENIX Security Symposium (USENIX Security 23), 1685-1702, 2023
472023
Can Adversarially Robust Learning Leverage Computational Hardness?
S Mahloujifar, M Mahmoody
[edit] Saeed Mahloujifar, Mohammad Mahmoody ; Proceedings of the 30th …, 2019
472019
Bounding training data reconstruction in dp-sgd
J Hayes, B Balle, S Mahloujifar
Advances in neural information processing systems 36, 78696-78722, 2023
432023
Membership inference on word embedding and beyond
S Mahloujifar, HA Inan, M Chase, E Ghosh, M Hasegawa
arXiv preprint arXiv:2106.11384, 2021
432021
Publicly-detectable watermarking for language models
J Fairoze, S Garg, S Jha, S Mahloujifar, M Mahmoody, M Wang
arXiv preprint arXiv:2310.18491, 2023
392023
Parameterizing activation functions for adversarial robustness
S Dai, S Mahloujifar, P Mittal
2022 IEEE Security and Privacy Workshops (SPW), 80-87, 2022
342022
Learning under -Tampering Attacks
S Mahloujifar, DI Diochnos, M Mahmoody
Proceedings of Algorithmic Learning Theory, PMLR 83:572-596, 2018., 2018
342018
Blockwise p-Tampering Attacks on Cryptographic Primitives, Extractors, and Learners
S Mahloujifar, M Mahmoody
Theory of Cryptography: 15th International Conference, TCC 2017, Baltimore …, 2017
312017
Experimenting with zero-knowledge proofs of training
S Garg, A Goel, S Jha, S Mahloujifar, M Mahmoody, GV Policharla, ...
Proceedings of the 2023 ACM SIGSAC Conference on Computer and Communications …, 2023
302023
Systemet kan ikke utføre handlingen. Prøv på nytt senere.
Artikler 1–20