Limitations of Stochastic Selection with Pairwise Independent Priors

Explore the limitations of stochastic selection problems with pairwise-independent priors and matroid constraints in this Google TechTalk presented by Neel Patel. Delve into the essentially-optimal bounds for contention resolution and prophet inequalities, examining both negative results for general matroids and positive outcomes for matroids with the partition property. Learn about the construction of pairwise-independent distributions and their impact on balanced offline CRS and competitive prophet inequalities. Discover how random linear maps transform linear independence into stochastic independence, and understand the implications for combinatorial optimization under uncertainty. Gain insights into the speaker's research interests, including algorithmic contract theory, computational economics, and dynamic algorithms for graph problems.