From Differential Privacy to Bounds on Membership Inference: Less can be More.
Trans. Mach. Learn. Res., 2024
Finding Optimally Robust Data Mixtures via Concave Maximization.
CoRR, 2024
Unlearnable Algorithms for In-context Learning.
CoRR, 2024
Gradients Look Alike: Sensitivity is Often Overestimated in DP-SGD.
Proceedings of the 33rd USENIX Security Symposium, 2024
Better Sparsifiers for Directed Eulerian Graphs.
Proceedings of the 51st International Colloquium on Automata, Languages, and Programming, 2024
Training Private Models That Know What They Don't Know.
Proceedings of the Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, 2023
Proof-of-Learning is Currently More Broken Than You Think.
Proceedings of the 8th IEEE European Symposium on Security and Privacy, 2023
On the Fundamental Limits of Formally (Dis)Proving Robustness in Proof-of-Learning.
CoRR, 2022
Selective Classification Via Neural Network Training Dynamics.
CoRR, 2022
Bounding Membership Inference.
CoRR, 2022
On the Necessity of Auditable Algorithmic Definitions for Machine Unlearning.
Proceedings of the 31st USENIX Security Symposium, 2022
Unrolling SGD: Understanding Factors Influencing Machine Unlearning.
Proceedings of the 7th IEEE European Symposium on Security and Privacy, 2022
SoK: Machine Learning Governance.
CoRR, 2021
Proof-of-Learning: Definitions and Practice.
Proceedings of the 42nd IEEE Symposium on Security and Privacy, 2021