no code implementations • ICML 2020 • Jinshuo Dong, David Durfee, Ryan Rogers
We consider precise composition bounds of the overall privacy loss for exponential mechanisms, one of the fundamental classes of mechanisms in DP.
no code implementations • 2 Nov 2023 • David Durfee
We provide a new algorithmic framework for differentially private estimation of general functions that adapts to the hardness of the underlying dataset.
no code implementations • 19 Feb 2023 • Kayhan Behdin, Qingquan Song, Aman Gupta, Sathiya Keerthi, Ayan Acharya, Borja Ocejo, Gregory Dexter, Rajiv Khanna, David Durfee, Rahul Mazumder
Modern deep learning models are over-parameterized, where different optima can result in widely varying generalization performance.
no code implementations • 7 Dec 2022 • Kayhan Behdin, Qingquan Song, Aman Gupta, David Durfee, Ayan Acharya, Sathiya Keerthi, Rahul Mazumder
To that end, this paper presents a thorough empirical evaluation of mSAM on various tasks and datasets.
no code implementations • 10 Feb 2022 • David Durfee, Aman Gupta, Kinjal Basu
We introduce the notion of heterogeneous calibration that applies a post-hoc model-agnostic transformation to model outputs for improving AUC performance on binary classification tasks.
1 code implementation • NeurIPS 2019 • David Durfee, Ryan M. Rogers
We design algorithms that ensures (approximate) differential privacy and only needs access to the true top-k' elements from the data for any chosen k' ≥ k. This is a highly desirable feature for making differential privacy practical, since the algorithms require no knowledge of the domain.
1 code implementation • 10 May 2019 • David Durfee, Ryan Rogers
We study the problem of top-$k$ selection over a large domain universe subject to user-level differential privacy.
Cryptography and Security
no code implementations • 6 May 2019 • David Durfee, Yu Gao, Anup B. Rao, Sebastian Wild
We give an algorithm to compute a one-dimensional shape-constrained function that best fits given data in weighted-$L_{\infty}$ norm.