I work at Meta AI, FAIR Lab.
I got my Ph.D. in the Department of Electrical and Computer Engineering at Princeton University, working with Prof. Peter Ramadge and Prof. Karthik Narasimhan between 2017 and 2022.
My research interests lie at the intersection of machine learning, reinforcement learning, and natural language processing. Specifically, I work on safe reinforcement learning, focusing on building autonomous systems that acquire knowledge by interacting with the world (multi-modal representation), and providing provable safety guarantees during training and deployment (safety and sample-efficiency). My work has been published and presented in leading machine learning venues, including NeurIPS, ICML, ICLR, and EMNLP.
I have been working on machine learning projects, including but not limited to
(1) Reinforcement learning with safety constraints for self-driving cars;
(2) Learning control policy for robotics systems;
(3) Anomaly detection from sensor measurements;
(4) Instruction following in natural language processing;
(5) Simulation to real transfer using physics priors and transformers;
(6) Representation learning from user behavioral data.
Before Princeton, I received my BS in Electrical Engineering and Computer Science (Honors Program) at National Chiao Tung University, Taiwan.
yangtsungyen [at] gmail.com / Google Scholar / LinkedIn / Twitter
News:
Sep 2022, One paper was accepted at NeurIPS 2022
Sep 2022, One paper was accepted at CoRL 2022
Jun 2022, One paper was accepted at IROS 2022, paper
Jun 2022, New preprint "Leveraging Language for Accelerated Learning of Tool Manipulation" paper
May 2022, The Google AI Blog post "Learning Locomotion Skills Safely in the Real World" is live now, twitter
Mar 2022, New preprint "Safe Reinforcement Learning for Legged Locomotion", paper, project website, TL;DR: We are able to learn locomotion skills without falling in the real world
Sep 2021, One paper was accepted at NeurIPS (Spotlight, 3% of overall submissions) 2021
May 2021, One paper was accepted at ICML 2021, paper
Nov 2020, Two papers was accepted at NeurIPS Workshop 2020 (One spotlight talk), paper, paper
Oct 2020, New preprint "Safe Reinforcement Learning with Natural Language Constraints" paper, demo
Sep 2020, Paper "Robust and Interpretable Grounding of Spatial References with Relation Networks" was accepted in Findings of EMNLP 2020, paper
Jun 2020, New preprint "Accelerating Safe Reinforcement Learning with Constraint-mismatched Policies" paper
Dec 2019, Paper "Projection-Based Constrained Policy Optimization" was accepted in ICLR 2020
Jan 2019, I was a recipient of the Siemens Fellowship
Experience:
Industry:
Researcher, Meta AI, Aug 2022 - Present, NYC, NY
Research Intern & Student Researcher, Google Brain, May 2021 - May 2022, Mountain View, CA
Research Intern, Siemens Research, May 2020 - Aug 2020, Princeton, NJ
Research Intern, Siemens Research, May 2018 - Aug 2018, Princeton, NJ
Academia:
Graduate Researcher, ECE Department and Princeton NLP Group, Princeton University, Jun 2017 - Present, Princeton, NJ
Academic Activities:
Reviewer:
NeurIPS, ICML, ACL, EMNLP, AAAI, CISS
Teaching:
SML201 - Introduction to data science
ECE435/COS434 - Machine learning theory
ECE435/535 - Machine learning and pattern recognition