Research map / Bandits and online learning / Online convex optimization and games
Private Online Prediction from Experts with Small Losses
Bo Li, Peng Ye
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Gives differentially private online prediction-from-experts algorithms whose regret scales with the best expert's cumulative loss L* rather than only with the horizon T.
Problem
Private online prediction from experts achieves near-optimal Õ(√T) regret, which is suboptimal when the best expert's cumulative loss L* is much smaller than T.
Approach
First differentially private small-loss (L*-dependent) regret bounds for online experts, improving on the standard private Õ(√T) guarantee when the best expert has small loss.
Experimental setup
Metrics: regret
Closest papers by meaning
- Private Online Learning via Lazy Algorithms (NeurIPS 2024 · similarity 0.92)
- Federated Online Prediction from Experts with Differential Privacy: Separations and Regret Speed-ups (NeurIPS 2024 · similarity 0.92)
- Tracking The Best Expert Privately (ICML 2025 · similarity 0.92)
- Faster Rates for Private Adversarial Bandits (ICML 2025 · similarity 0.88)
- Improved Differentially Private and Lazy Online Convex Optimization: Lower Regret without Smoothness Requirements (ICML 2024 · similarity 0.88)
- The Limits of Differential Privacy in Online Learning (NeurIPS 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
