Hi! I'm a second-year CS PhD student studying machine learning at Stanford University, advised by Chelsea Finn. My research is funded by a National Science Foundation Graduate Fellowship. I also spend one day a week applying machine learning to systems problems at Google Research. Previously, I spent a year as a Google AI resident and before that, completed my master's (CS) and my undergraduate degrees (CS and math) at Stanford, advised by Percy Liang.
My goal is to build practical agents that can help humans. Since it's often infeasible to directly train agents on the environments they may be deployed, one key aspect of practical agents is the ability to quickly adapt to new environments. For example, it's infeasible to train a robot chef directly in a home kitchen, but it is possible to train the robot on many diverse kitchens in a robot factory. Toward this end, I study the problem of meta-exploration: how to best explore a new environment in order to quickly adapt to it, e.g., by finding ingredients in a new kitchen (ArXiv 2020).
I also work toward practical agents by applying machine learning to computer systems, where data is plentiful and advances can yield tangible real-world impact. Here, I aim to replace handcrafted heuristics with learned models, which can significantly improve performance, e.g., in cache replacement (ICML 2020). Of course, directly deploying these learned models requires solving additional challenges (e.g., minimizing model latency and size), but in the meantime, these models can still improve existing systems (e.g., by yielding insights that can be incorporated through more traditional avenues). I am also fascinated by computer security and would love to apply reinforcement learning ideas to automatically detecting vulnerabilities.
I love rock climbing and spend most of my free time bouldering indoors, although I'm hoping to climb more outside in the future. I'm always on the lookout for more climbing friends, so let me know if you like to climb or want to try it out!
You can reach me at [my first name][my last name]@cs.stanford.edu
Last updated: September 10, 2020