Algorithmic Learning Theory and Differential Privacy - Session 3

Explore cutting-edge research in algorithmic learning theory through this conference session from the 37th International Conference on Algorithmic Learning Theory (ALT 2026) and ShaiFest. Delve into six research presentations covering advanced topics in statistical learning, privacy-preserving algorithms, and optimization theory. Learn about closeness testing methodologies for distributed measurement systems and discover nearly minimax approaches to discrete distribution estimation using Kullback-Leibler divergence with high probability guarantees. Examine purely private covariance estimation techniques and understand the complexities of differentially private bilevel optimization problems. Investigate private learning algorithms for decision lists and explore differentially private versions of the Winnow algorithm. Conclude with insights into improved regret bounds for stochastic decision-theoretic online learning under differential privacy constraints, presented by leading researchers including Clement Louis Canonne, Aditya Vikram Singh, Dirk van der Hoeven, Julia Olkhovskaya, Tim van Erven, Tommaso d'Orsi, Gleb Novikov, Guy Kornowski, Mark Bun, William Fang, Ruihan Wu, and Yu-Xiang Wang.