Moshe Shenfeld

Moshe Shenfeld

Built with Academic Pages

My full publications list can be found on my Google Scholar profile, alongside a bunch of patents.

Proceedings

Near-Optimal Private Linear Regression via Iterative Hessian Mixing

With: Lev, Omri and Srinivasan, Vishwak and Ligett, Katrina and Wilson, Ashia C

Accepted to the 43rd International Conference on Machine Learning (ICML 2026) (Spotlight)

arXiv

Efficient computation of the privacy loss distribution for random allocation

With: Vitaly Feldman

Accepted to the 43rd International Conference on Machine Learning (ICML 2026)

arXiv

The Gaussian Mixing Mechanism: Renyi Differential Privacy via Gaussian Sketches

With: Omri Lev, Vishwak Srinivasan, Katrina Ligett, Ayush Sekhari, and Ashia C Wilson

Accepted to the 38th Advances in Neural Information Processing Systems (NeurIPS 2025)

arXiv

Privacy amplification by random allocation

With: Vitaly Feldman

Accepted to the 38th Advances in Neural Information Processing Systems (NeurIPS 2025) (Spotlight)

arXiv

How Well Can Differential Privacy Be Audited in One Run?

With: Amit Keinan and Katrina Ligett

Accepted to the 38th Advances in Neural Information Processing Systems (NeurIPS 2025) (Spotlight)

arXiv Talk

Generalization in the face of adaptivity: A bayesian perspective

With: Katrina Ligett

36th Advances in Neural Information Processing Systems (NeurIPS 2023) (Spotlight)

Conference arXiv Talk

A new analysis of differential privacy's generalization guarantees

With: Christopher Jung, Katrina Ligett, Seth Neel, Aaron Roth, and Saeed Sharifi-Malvajerdi

11th Innovations in Theoretical Computer Science Conference (ITCS 2020)

Conference arXiv

A necessary and sufficient stability notion for adaptive generalization

With: Katrina Ligett

32nd Advances in Neural Information Processing Systems (NeurIPS 2019)

Conference arXiv Talk

Workshop Papers

Differentially Private Non-Parametric Confidence Intervals

With: Katrina Ligett, Tomer Shoham, and Noa Velner-Harris

Theory and Practice of Differential Privacy (TPDP 2025)

Conference

Preprints

The Fast Mixing Mechanism for Differential Privacy

With: Lev, Omri and Shenfeld, Moshe and Srinivasan, Vishwak and Ligett, Katrina and Wilson, Ashia C

arXiv preprint arXiv:2605.30600 , 2026

arXiv

Black-Box Differentially Private Nonparametric Confidence Intervals Under Minimal Assumptions

With: Shoham, Tomer and Shenfeld, Moshe and Velner-Harris, Noa and Ligett, Katrina

arXiv preprint arXiv:2511.01303 , 2025

arXiv

Publications in other fields

The Right to Education

With: Shenfeld, Moshe

Off the Derech: Leaving Orthodox Judaism

Conference