UCSB CS Theory Colloquium Series

Fall 2021

Friday, October 1, 2pm, Phelps 2510
                Speaker: Grant Schoenebeck (Michigan)
                Title: Eliciting Expert Information without Verification
Abstract:
A central question of crowd-sourcing is how to elicit expertise from agents. This is even more difficult when answers cannot be directly verified. A key challenge holding back this field is that sophisticated agents may strategically withhold effort or information when they believe their payoff will be based upon comparison with other agents whose reports will likely omit this information due to lack of effort or expertise.

This talk will argue that information theory provides the “right way” to think about these problems in that measuring information theoretic properties can directly translate into incentive compatible mechanisms. Moreover, this talk will show how a soft-predictor for an agent's report (given the other agents' reports) can typically be leveraged to provide measurements of the relevant information theoretic properties (thus yielding incentive compatible mechanisms).