Saeed Mahloujifar

Luca Melis

Kamalika Chaudhuri

CoRR, 2024

Aligning LLMs to Be Robust Against Prompt Injection.

[BibT_eX]

[DOI]

CoRR, 2024

Guarantees of confidentiality via Hammersley-Chapman-Robbins bounds.

[BibT_eX]

[DOI]

Kamalika Chaudhuri

Chuan Guo

Laurens van der Maaten

Mark Tygert

CoRR, 2024

Privacy Amplification for the Gaussian Mechanism via Bounded Support.

[BibT_eX]

[DOI]

CoRR, 2024

Private Fine-tuning of Large Language Models with Zeroth-order Optimization.

[BibT_eX]

[DOI]

CoRR, 2024

Horus: Granular In-Network Task Scheduler for Cloud Datacenters.

[BibT_eX]

[DOI]

Proceedings of the 21st USENIX Symposium on Networked Systems Design and Implementation, 2024

A New Linear Scaling Rule for Private Adaptive Hyperparameter Optimization.

[BibT_eX]

[DOI]

Proceedings of the Forty-first International Conference on Machine Learning, 2024

2023

Experimenting with Zero-Knowledge Proofs of Training.

[BibT_eX]

[DOI]

Guru-Vamsi Policharla

Mingyuan Wang

IACR Cryptol. ePrint Arch., 2023

Publicly Detectable Watermarking for Language Models.

[BibT_eX]

[DOI]

IACR Cryptol. ePrint Arch., 2023

A Randomized Approach for Tight Privacy Accounting.

[BibT_eX]

[DOI]

CoRR, 2023

Towards A Proactive ML Approach for Detecting Backdoor Poison Samples.

[BibT_eX]

[DOI]

Proceedings of the 32nd USENIX Security Symposium, 2023

ObjectSeeker: Certifiably Robust Object Detection against Patch Hiding Attacks via Patch-agnostic Masking.

[BibT_eX]

[DOI]

Proceedings of the 44th IEEE Symposium on Security and Privacy, 2023

A Randomized Approach to Tight Privacy Accounting.

[BibT_eX]

[DOI]

Proceedings of the Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, 2023

Bounding training data reconstruction in DP-SGD.

[BibT_eX]

[DOI]

Jamie Hayes

Borja Balle

Proceedings of the Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, 2023

Uncovering Adversarial Risks of Test-Time Adaptation.

[BibT_eX]

[DOI]

Proceedings of the International Conference on Machine Learning, 2023

Effectively Using Public Data in Privacy Preserving Machine Learning.

[BibT_eX]

[DOI]

Proceedings of the International Conference on Machine Learning, 2023

MultiRobustBench: Benchmarking Robustness Against Multiple Attacks.

[BibT_eX]

[DOI]

Proceedings of the International Conference on Machine Learning, 2023

Revisiting the Assumption of Latent Separability for Backdoor Defenses.

[BibT_eX]

[DOI]

Proceedings of the Eleventh International Conference on Learning Representations, 2023

2022

Machine Learning with Differentially Private Labels: Mechanisms and Frameworks.

[BibT_eX]

[DOI]

Proc. Priv. Enhancing Technol., 2022

DP-RAFT: A Differentially Private Recipe for Accelerated Fine-Tuning.

[BibT_eX]

[DOI]

CoRR, 2022

Overparameterized (robust) models from computational constraints.

[BibT_eX]

[DOI]

CoRR, 2022

Fight Poison with Poison: Detecting Backdoor Poison Samples via Decoupling Benign Correlations.

[BibT_eX]

[DOI]

CoRR, 2022

Circumventing Backdoor Defenses That Are Based on Latent Separability.

[BibT_eX]

[DOI]

CoRR, 2022

Optimal Membership Inference Bounds for Adaptive Composition of Sampled Gaussian Mechanisms.

[BibT_eX]

[DOI]

Alexandre Sablayrolles

Graham Cormode

Somesh Jha

CoRR, 2022

Mitigating Membership Inference Attacks by Self-Distillation Through a Novel Ensemble Architecture.

[BibT_eX]

[DOI]

Proceedings of the 31st USENIX Security Symposium, 2022

PatchCleanser: Certifiably Robust Defense against Adversarial Patches for Any Image Classifier.

[BibT_eX]

[DOI]

Chong Xiang

Prateek Mittal

Proceedings of the 31st USENIX Security Symposium, 2022

Parameterizing Activation Functions for Adversarial Robustness.

[BibT_eX]

[DOI]

Sihui Dai

Prateek Mittal

Proceedings of the 43rd IEEE Security and Privacy, 2022

Renyi Differential Privacy of Propose-Test-Release and Applications to Private and Robust Machine Learning.

[BibT_eX]

[DOI]

Proceedings of the Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems 2022, 2022

Overparameterization from Computational Constraints.

[BibT_eX]

[DOI]

Proceedings of the Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems 2022, 2022

Formulating Robustness Against Unforeseen Attacks.

[BibT_eX]

[DOI]

Sihui Dai

Prateek Mittal

Proceedings of the Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems 2022, 2022

Robust Learning Meets Generative Models: Can Proxy Distributions Improve Adversarial Robustness?

[BibT_eX]

[DOI]

Proceedings of the Tenth International Conference on Learning Representations, 2022

Just Rotate it: Deploying Backdoor Attacks via Rotation Transformation.

[BibT_eX]

[DOI]

Proceedings of the 15th ACM Workshop on Artificial Intelligence and Security, 2022

SparseFed: Mitigating Model Poisoning Attacks in Federated Learning with Sparsification.

[BibT_eX]

[DOI]

Proceedings of the International Conference on Artificial Intelligence and Statistics, 2022

2021

Polynomial-time targeted attacks on coin tossing for any number of corruptions.

[BibT_eX]

[DOI]

IACR Cryptol. ePrint Arch., 2021

Property Inference from Poisoning.

[BibT_eX]

[DOI]

Melissa Chase

Esha Ghosh

IACR Cryptol. ePrint Arch., 2021

NeuraCrypt is not private.

[BibT_eX]

[DOI]

CoRR, 2021

Membership Inference on Word Embedding and Beyond.

[BibT_eX]

[DOI]

CoRR, 2021

Improving Adversarial Robustness Using Proxy Distributions.

[BibT_eX]

[DOI]

CoRR, 2021

Is Private Learning Possible with Instance Encoding?

[BibT_eX]

[DOI]

Proceedings of the 42nd IEEE Symposium on Security and Privacy, 2021

A Separation Result Between Data-oblivious and Data-aware Poisoning Attacks.

[BibT_eX]

[DOI]

Abhradeep Guha Thakurta

Proceedings of the Advances in Neural Information Processing Systems 34: Annual Conference on Neural Information Processing Systems 2021, 2021

Model-Targeted Poisoning Attacks with Provable Convergence.

[BibT_eX]

[DOI]

Proceedings of the 38th International Conference on Machine Learning, 2021

2020

An Attack on InstaHide: Is Private Learning Possible with Instance Encoding?

[BibT_eX]

[DOI]

CoRR, 2020

Model-Targeted Poisoning Attacks: Provable Convergence and Certified Bounds.

[BibT_eX]

[DOI]

CoRR, 2020

Obliviousness Makes Poisoning Adversaries Weaker.

[BibT_eX]

[DOI]

CoRR, 2020

Learning under p-tampering poisoning attacks.

[BibT_eX]

[DOI]

Ann. Math. Artif. Intell., 2020

Computational Concentration of Measure: Optimal Bounds, Reductions, and More.

[BibT_eX]

[DOI]

Omid Etesami

Proceedings of the 2020 ACM-SIAM Symposium on Discrete Algorithms, 2020

Lower Bounds for Adversarially Robust PAC Learning.

[BibT_eX]

[DOI]

Proceedings of the International Symposium on Artificial Intelligence and Mathematics, 2020

Lower Bounds for Adversarially Robust PAC Learning under Evasion and Hybrid Attacks.

[BibT_eX]

[DOI]

Proceedings of the 19th IEEE International Conference on Machine Learning and Applications, 2020

Adversarially Robust Learning Could Leverage Computational Hardness.

[BibT_eX]

[DOI]

Proceedings of the Algorithmic Learning Theory, 2020

2019

Empirically Measuring Concentration: Fundamental Limits on Intrinsic Robustness.

[BibT_eX]

[DOI]

Proceedings of the Advances in Neural Information Processing Systems 32: Annual Conference on Neural Information Processing Systems 2019, 2019

Data Poisoning Attacks in Multi-Party Learning.

[BibT_eX]

[DOI]

Ameer Mohammed

Proceedings of the 36th International Conference on Machine Learning, 2019

Can Adversarially Robust Learning LeverageComputational Hardness?

[BibT_eX]

[DOI]

Proceedings of the Algorithmic Learning Theory, 2019

The Curse of Concentration in Robust Learning: Evasion and Poisoning Attacks from Concentration of Measure.

[BibT_eX]

[DOI]

Proceedings of the Thirty-Third AAAI Conference on Artificial Intelligence, 2019

2018

Multi-party Poisoning through Generalized p-Tampering.

[BibT_eX]

[DOI]

Ameer Mohammed

IACR Cryptol. ePrint Arch., 2018

Can Adversarially Robust Learning Leverage Computational Hardness?

[BibT_eX]

[DOI]

CoRR, 2018

Adversarial Risk and Robustness: General Definitions and Implications for the Uniform Distribution.

[BibT_eX]

[DOI]

Proceedings of the Advances in Neural Information Processing Systems 31: Annual Conference on Neural Information Processing Systems 2018, 2018

Learning under $p$-Tampering Attacks.

[BibT_eX]

[DOI]

Proceedings of the Algorithmic Learning Theory, 2018

2017

Blockwise p-Tampering Attacks on Cryptographic Primitives, Extractors, and Learners.

[BibT_eX]

[DOI]