(Sequential) Information Design

Published on ● Video Link: https://www.youtube.com/watch?v=OHBOJTYxAR0



Duration: 1:02:00
1,848 views
0


Laura Doval (Columbia University)
https://simons.berkeley.edu/talks/tbd-452
Data-Driven Decision Processes Boot Camp

To summarize the possible outcomes of a strategic interaction, after having specified the players’ actions and payoffs, an analyst makes assumptions about their information and the extensive form that ties all of these elements together. Conventional equilibrium concepts begin by fixing both and proceed to provide a set of solutions for that fully specified game. An analyst, however, may know the payoff-relevant data but not the players’ private information, nor the extensive form that governs their play. Alternatively, a designer may be able to build a mechanism from these ingredients. The talk will survey the information design literature in Economics, which provides solution concepts that either fix the extensive-form game, but not the information structure (Bayes' correlated equilibrium), or do not fix either (coordinated equilibrium). Along the way, we will discuss applications to Economics, Computer Science, and Operations.







Tags:
Simons Institute
theoretical computer science
UC Berkeley
Computer Science
Theory of Computation
Theory of Computing
Data-Driven Decision Processes Boot Camp
Laura Doval