Jason (Yecheng) Ma

I am a PhD student at the University of Pennsylvania advised by Osbert Bastani and Dinesh Jayaraman. I work on reinforcement learning with a particular interest in safety, model-based RL, and generalization.

During my undergraduate studies at Harvard, I worked with David Parkes on imitation learning. I received the highest honor distinction in Computer Science, and was a winner of the Thomas Temple Hoopes Prize. Before AI research, I did quantitative finance in summer internships.

Google Scholar  /  CV  /  Email

  • October 2021: New pre-print on safe MBRL accepted to two NeurIPS workshops!
  • September 2021: One NeurIPS paper on offline distributional RL accepted!
  • July 2021: One ICCV paper on diverse trajectory forecasting accepted!
  • May 2021: One ICML paper on robust off-policy evaluation accepted!
  • May 2021: One RSS/ICML workshop paper on safe human-in-the-loop planning accepted!
  • May 2020: Graduated from Harvard College with Highest Honors and Hoopes Prize!
Research (Highlighted / All)

Inspired by this and this template. Hosted on Eniac.