Wen Sun, assistant professor of computer science in the Cornell Ann S. Bowers College of Computing and Information Science, has received the National Science Foundation’s Faculty Early Career Development (NSF CAREER) award, which supports early-career faculty who have the potential to advance their academic fields.
Sun will receive $600,000 over the next five years to support his research in reinforcement learning, a type of machine learning algorithm that can enable intelligent systems to learn and improve from their own experience. It’s a technique Sun believes is critical for achieving superhuman machine intelligence. Specifically, Sun will develop new reinforcement learning algorithms that have strong theoretical guarantees and are both scalable to big data and applicable to modern artificial intelligence (AI) applications. These algorithms would have broad impact, helping to optimize database query time, develop AI-powered network security “cyberagents”, and improve generative AI.
Most recently, Sun and his research group have used reinforcement learning to fine-tune generative models – like large language models (LLMs, e.g., ChatGPT) and diffusion models – using human feedback, and developed new reinforcement learning algorithms that can learn from human interaction and learn jointly from static offline data and online interactive data.
Prior to joining the Cornell Bowers CIS faculty in 2020, Sun was a postdoctoral researcher at Microsoft Research NYC, where he continues to serve as a part-time researcher. He received a Ph.D. in robotics from Carnegie Mellon University in 2019. Sun’s past research is supported by awards from NSF, Defense Advanced Research Projects Agency (DARPA), and Infosys.