About Me

Hi, I’m Yifan!

I’m a final-year PhD student at Northwestern University, advised by Prof. Jason Hartline. I received my B.S. in Computer Science from Turing Class at Peking University in 2020, where I was advised by Prof. Yuqing Kong.

I have a broad interest in the intersections between theoretical computer science, AI, and economics. Recently, I focus on building a decision-theoretic framework for trustworthy AI.

My Research

In the modern AI workflow, upstream AI predictions are provided as a service to downstream decision-making instead of designed for an application. My research evaluates predictors and designs algorithms out of the context of a particular application, through statistical decision theory that models rational decisions under uncertainty. With the abstraction of a decision problem in statistical decision theory, my framework aims to achieve simultaneous guarantee for every decision-maker, aligning closely with the modern workflow without a specific decision problem.

At the core of my techniques is proper scoring rules, the class of functions that evaluate a probabilistic prediction by the decision payoff it leads to. My research focuses on the evaluation of trustworthy AI-assisted decision-making and algorithm design that enables trustworthiness. Currently, my research topics include

• optimal information elicitation (optimization of proper scoring rules), a theoretical foundation to my framework;
• calibration, trustworthy predictions for all decision-makers;
• benchmarks for human-AI interaction, theoretically justified framework for understanding human behavior in AI-assisted decision problems.