AI Paper Atlas

Research map / Reinforcement learning / MDPs and planning

Loss Dynamics of Temporal Difference Reinforcement Learning

Blake Bordelon, Paul Masset, Henry Kuo, Cengiz Pehlevan

NeurIPS 2023 · Reinforcement learning › MDPs and planning · approach: POMDPs and confounded RL · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
increase batch size BTD learning with linear features · synthetic power-law MDPvalue error plateauImproved
increase discount factor gammaTD learning · synthetic power-law MDPvalue error plateauHurt
increase learning rate etaTD learning · synthetic power-law MDPvalue error plateauHurt
anneal learning rate as eta_0 n^-chiTD learning · synthetic power-law MDPvalue errorImproved
use very large annealing exponent chiTD learning · synthetic power-law MDPvalue errorHurt
scale-based reward shaping (w_phi parallel to w_TD)TD learning · synthetic MDPvalue error dynamicsNo clear effect
rotation-based reward shaping toward high-variance feature directionsTD learning · synthetic MDPconvergence timescaleImproved
dense reward instead of sparse rewardTD learning with place cell code · 2D exploration MDPvalue error convergenceImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy