Episode 25: Nicklas Hansen, UCSD, on long-horizon planning and why algorithms don't drive research progress
Generally Intelligent
00:00
Introduction
Nicholas Hansen is a PhD student at UC San Diego working with professors Shalom Wang and Hausu. He's interested in how to get reinforcement learning agents to generalize and adapt flexibly to new environments. "You can much easier fit that using a model than predicting all the possible futures using like pixel prediction," he says.
Transcript
Play full episode
