Meta Gradients in Learning
Tom explains how meta gradients help balance non-stationarity in learning dynamics, especially in scenarios like constrained MDPs. The discussion sheds light on the benefits of meta policy gradients in dealing with non-stationary problems.In this clip
From this podcast
Machine Learning Street Talk (MLST)
#49 - Meta-Gradients in RL - Dr. Tom Zahavy (DeepMind)
Related Questions