Can we build a generalist agent? Dr. Minqi Jiang and Dr. Marc Rigter
Machine Learning Street Talk (MLST)
00:00
Navigating Minimax Regret in Machine Learning
This chapter explores curriculum learning and machine teaching, focusing on the concept of minimax regret as a robust objective for training agents. It contrasts model-based reinforcement learning with value-based methods, highlighting the importance of optimizing for suboptimality in diverse environments.
Transcript
Play full episode
