I’m a first-year Ph.D. student in Economic Analysis & Policy group at Stanford Graduate School of Business (GSB) with PhD minor in Electrical Engineering. My PhD fields are in Advanced Theory, Advanced Econometrics, and Market Design.
Previously, I received an M.S. degree in Computational and Mathematical Engineering (ICME) and a B.S. degree in Symbolic Systems (with honors) from Stanford University. I am honored to have Prof. Yuliy Sannikov as the advisor and Prof. Robert Wilson as the second reader for my undergraduate thesis.
Research Interests
Economic Theory
Working Papers
Score-Debiased Kernel Density Estimation, accepted for ICLR Workshop 2025 (Frontiers in Probabilistic Inference).
Thanawat Sornwanee, Elliot Epstein, Rajat Dwaraknath, Jerry Liu, and John Winnicki.
Abstract: We propose a novel method for density estimation that leverages an estimated score function to debias kernel density estimation (SD-KDE). In our approach, each data point is adjusted by taking a single step along the score function with a specific choice of step size, followed by standard KDE with a modified bandwidth. The step size and modified bandwidth are chosen to remove the leading order bias in the KDE. Our experiments on synthetic tasks in 1D, 2D and on MNIST, demonstrate that our proposed SD-KDE method significantly reduces the mean integrated squared error compared to the standard Silverman KDE, even with noisy estimates in the score function. These results underscore the potential of integrating score-based corrections into nonparametric density estimation. Reinforced Social Learning, presented at INFORMS Annual Meeting 2024.
Thanawat Sornwanee.
Abstract: I present a knowledge-free mechanism to ensure the zero asymptotic average regret in a sequential social learning framework with binary state of the world and homogeneous preference. This mechanism also ensures that the state of the world can be asymptotically learned, regardless of the choice of Nash equilibrium employed. The asymptotic optimality of the mechanism will also shown to be preserved under different economic scenarios, such as when each agent may be blinded from the history of actions and/or received a utility shock.Teaching & Services
Summer 2023: Workshop Assistant: Stanford ICME Workshop, Data Visualization
Autumn 2022: Teaching Assistant: Stanford MS&E 339: Algorithms for Decntralized Finance
Grading:
- Stanford MGTECON 200: Managerial Economics
- Stanford ECON 202: Microeconomics I
- Stanford MS&E 232: Introduction to Game Theory
- Stanford EE 392F: Large-Scale Convex Optimization: Algorithms and Analyses via Monotone Operators
Awards
- Gold Medal, International Physics Olympiad (50th IPhO, Tel Aviv, Israel) [2019]
- Honorable Mentions, Asian Earth Science Olympiad (18th & 19th APhO, Yakutsk, Russia & Hanoi, Vietnam) [2017, 2018]
- Bronze Medal, International Earth Science Olympiad (10th IESO, Mie, Japan) [2016]
Talks & Presentations
2025
- “Score-Debiased Kernel Density Estimation” ICLR 2025 Workshop, Frontiers in Probabilistic Inference: Sampling Meets Learning, Singapore
- “Score-Debiased Kernel Density Estimation” Workshop on Experimental Design: AI for Science, Stanford, California
2024
- “Reinforced Social Learning” INFORMS Annual Meeting 2024, Seattle, Washington
Contact
Email: tsornwanee [at] stanford.edu