When: November 12, 2014, 12:30 PM–1:30 PM
Quentin Berthet from Caltech shows how the hypotheses from theoretical computer science on the hardness of some problems in average case can give a better understanding of algorithmic limitations for high-dimensional statistical problems. Several results related to the planted clique problem are shown, and a problem is studied involving random satisfiability formulas.
SILO is a lecture series organized through WID’s Optimization group with speakers from the UW faculty, graduate students or invited researchers that discuss mathematical related topics. SILO’s purpose is to provide a forum that helps connect and recruit mathematically-minded graduate students. SILO is a lunch-and-listen format, where speakers present interesting math topics while the audience eats lunch.