Research Interests
I have seen very little research. But in the little that I have seen, I find that I'm interested in research topics that both involve interesting math and have some direct applications to the world.
In the last two years, I worked on approximation algorithms for problems in operations research. (For my past work, see Publications.) Currently I'm working on problems in computational economics and mechanism design. I'm intending to pursue graduate school in computer science after graduating from Duke.
Current Projects
I'm currently working on the following projects. (Please contact me if you have any insights on any of these topics.)
Near Optimal Bayesian Auction under Private Budgets:
In 1982, Myerson characterized the revenue-maximizing Bayesian auction when bidders only have private valuations. However, little is currently known about the case when bidders also have private budget constraints. We seek to construct algorithms for cases that are polynomial-time implementable, and prove hardness results for cases that are not. (This work is joint with my work-study advisor Kamesh Munagala.)
Information Elicitation Mechanisms:
We study mechanisms for eliciting information in a variety of settings. Examples of such mechanisms include proper scoring rules, prediction markets, and feedback reporting schemes. We seek mechanisms that satisfy other desirable properties. For example, we are studying prediction mechanisms that do not incentivize undesirable actions (i.e. a prediction market on whether a terrorist attack will occur, without providing incentives for someone to bet for an attack and commit the attack.) Another example is prediction mechanisms that incentivize a certain amount of effort, under natural models for "effort." We are also studying feedback reporting schemes that incentivize users of a product to honestly report their experience, while guarding against undesirable equilibria. (This work is joint with my C-SURF advisor Vincent Conitzer.)