Hi, I’m Kaiwen. I seek to understand natural and artificial intelligence, then design systems that harness both.
I am a research staff at Enigma, Stanford University, and I recently graduated from UC San Diego. I am broadly interested in representation learning: how intelligence emerges from the representations it builds of the world, and how autonomous systems can use them to make discoveries on their own.
During my undergrad, I was fortunate to be advised by Talmo D. Pereira at the Salk Institute, Scott W. Linderman at Stanford University, and Yusu Wang and Gal Mishne at the Halıcıoğlu Data Science Institute.
News
Research Experience
Research Staff
Building brain foundation models and harvesting natural intelligence.
Research Intern
Developing computationally efficient deep RL imitation systems (“MIMIC-MJX: Neuromechanical Emulation of Animal Behavior”) for bio-mechanically realistic agents, discovering topological structure in embodied behavior spaces (“Shaped by What's Missing: Topological Invariance Simplification Discovers Behavioral Transitions”), and building code-to-action embodied agents (“Discrete Actions for Naturalistic Behavior in Embodied Biomechanical Animal Models”).
Visiting Research Scholar
Creating latent dynamical models for planning and generating bio-mechanically realistic behaviors using deep state-space modeling methods. Co-advised on the “Discrete Actions for Naturalistic Behavior in Embodied Biomechanical Animal Models” embodied agent project from Salk.
B.S. Data Science
B.S. Cognitive Behavioral Neuroscience
Graph compositional abstraction for molecular generation (“Beyond Flat Walks: Compositional Abstraction for Autoregressive Graph Generation”) and topological analysis of embodied agent behaviors. Co-advised on the “Shaped by What's Missing: Topological Invariance Simplification Discovers Behavioral Transitions” project from Salk.