Martin Pelikan

Researcher, Apple

1 paper at NeurIPS 2025

Homepage· OpenReview· Semantic Scholar· Google Scholar

Poster Session 3

Thursday, December 4, 2025 · 11:00 AM → 2:00 PM

Enabling Differentially Private Federated Learning for Speech Recognition: Benchmarks, Adaptive Optimizers, and Gradient Clipping

#114 · Martin Pelikan, Sheikh Shams Azam, Vitaly Feldman, Jan Silovsky, Kunal Talwar, Christopher Brinton, Tatiana Likhomanenko

We introduce the first practical benchmark for FL with DP in ASR, combining theoretical insights on gradient heterogeneity with empirical results that demonstrate scalability and strong performance under user-level privacy guarantees.