I'm now third-year. My research area is mathematical modeling of higher-order cognition, including Decision-making, Problem-solving and Learning, etc.

I'm currently working on the "wisdom of crowds" in decision tasks, e.g. bandit problems, that is, to estimate and evaluate the decision-making strategy at the group level. I'm also involved in Rich Shiffrin's research on rational game solutions.

In the past, I have worked on the projects listed below:

1)Computational models of the bandit problems, with Michael Lee and Mark Steyvers.

One line of my research is to understand human decision-making under uncertainty. We develop simple, psychologically interpretable heuristics which address the trade-off between exploration and exploitation and compare them to the optimal benchmarks from machine learning theories.

From an experimenter's perspective, however, another interesting problem is to optimize the experimental design which will best discriminate between quantitative models, e.g. to select parameters of the environmental distributions. We solve this problem by adopting simulation-based Bayesian design framework that uses Markov Chain Monte Carlo methods.

2)Mathematical models and statistics for experimental psychology, with Geoff Iverson.

Following Psychological Science's formal introduction of the statistic P_rep (Probability of replication) in lieu of the traditional P-value, we try to clarify the definition, meaning, calculation and interpretation of this statistic from a Bayesian perspective.