Abstract: How does a research community identify its blind spots—those problems that the community could be uniquely positioned to solve, or those methods that the community could apply to its existing problems, if only it knew of those relevant problems and methods that it was overlooking? How do two or more such communities connect to discover new research applications with social impact, well-motivated theoretical problems, and provide opportunities for people newly introduced to a field to grow into experts?
In this talk, I’ll instantiate these more general questions by discussing our ongoing collaborative work to connect researchers in mechanism design with students in sub-Saharan Africa. I’ll describe the work we’ve done in areas such as educational outreach, crowdsourcing contests, remote research collaborations, and preliminary ideas for market-based solutions. I’ll highlight our successes and failures thus far, speculate on promising directions to scale and generalize these efforts, and describe how you can help.
Speaker: Eric Sodomka, Research Scientist Manager, Facebook
Bio: Eric Sodomka manages the Algorithmic Game Theory and Market Design Research group within the Core Data Science team at Facebook. His group conducts research in mechanism design, optimization, and machine learning to solve practical problems related to ad auctions, community integrity, and social good. He received his PhD in computer science from Brown University, where he focused on sequential decision making and bidding in complex auctions.