A Theoretical Understanding of Gradient Bias in Meta-Reinforcement Learning

Nov 28, 2022

Speakers

About

Gradient-based Meta-RL (GMRL) refers to methods that maintain two-level optimisation procedures wherein the outer-loop meta-learner guides the inner-loop gradient-based reinforcement learner to achieve fast adaptations. In this paper, we develop a unified framework that describes variations of GMRL algorithms and points out that existing stochastic meta-gradient estimators adopted by GMRL are actually biased. Such meta-gradient bias comes from two sources: 1) the compositional bias incurred by the two-level problem structure, which has an upper bound of 𝒪(Kα^Kσ̂_In|τ|^-0.5) w.r.t. inner-loop update step K, learning rate α, estimate variance σ̂^2_In and sample size |τ|, and 2) the multi-step Hessian estimation bias Δ̂_H due to the use of autodiff, which has a polynomial impact 𝒪((K-1)(Δ̂_H)^K-1) on the meta-gradient bias. We study tabular MDPs empirically and offer quantitative evidence that testifies our theoretical findings on existing stochastic meta-gradient estimators. Furthermore, we conduct experiments on Iterated Prisoner's Dilemma and Atari games to show how other methods such as off-policy learning and low-bias estimator can help fix the gradient bias for GMRL algorithms in general.

Organizer

Like the format? Trust SlidesLive to capture your next event!

Professional recording and live streaming, delivered globally.

Sharing

Recommended Videos

Presentations on similar topic, category or speaker

Interested in talks like this? Follow NeurIPS 2022