Scholar
Dylan J. Foster
Google Scholar ID: RqwU8xsAAAAJ
Principal Researcher, Microsoft Research
Citations & Impact
Citations
5,791
H-index
37
i10-index
58
Publications
20
Co-authors
74
list available
Publications
Resume (English only)
Academic Achievements
- Selected papers include 'The Coverage Principle: How Pre-Training Enables Post-Training', 'Self-Improvement in Language Models: The Sharpening Mechanism', etc.; involved in organizing several workshops and courses such as the course 'Statistical Reinforcement Learning and Decision Making' at MIT.
Research Experience
- Principal Researcher at Microsoft Research, New England (and New York City), part of the Reinforcement Learning Group. Previously, a postdoctoral fellow at MIT Institute for Foundations of Data Science in IDSS.
Education
- Ph.D. in Computer Science from Cornell University (2019), advised by Karthik Sridharan; BS and MS in Electrical Engineering from the University of Southern California (2014).
Background
- Research interests: mathematical foundations—algorithm design principles and fundamental limits—necessary to develop intelligent agents that learn from experience. Currently most excited about the statistical and computational foundations of interactive decision making, including reinforcement learning and imitation learning; understanding and improving foundation models.
Miscellany
- Personal interests not mentioned.
