We had a delightful conversation covering: ⏺️ how I chose my PhD topic, ⏺️ connections with cognitive psychology and neuroscience, ⏺️ why I'm applying RL/AI in space today, ⏺️ what got me interested in machine learning in the first place, and more! Thank you, @robinchauhan.bsky.social!
E62: Abhishek Naik on Continuing RL & Average Reward How should RL handle non-episodic tasks, like Mars rovers or HPC scheduling? @abhisheknaik96.bsky.social shares insights from his PhD with Rich Sutton, plus how almost every discounted-reward algorithm can be improved by reward centering.