AI Paper Atlas

Research map / Reinforcement learning / Policy gradient and actor-critic

Distributional Policy Evaluation: a Maximum Entropy approach to Representation Learning

Riccardo Zamboni, Alberto Maria Metelli, Marcello Restelli

NeurIPS 2023 · Reinforcement learning › Policy gradient and actor-critic · approach: Distributional value learning · 0 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 boosting parameter beta from 3 to 450Distributional Max-Ent Progressive Factorization · GridWorld Block-MDP, N=50 and N=1000number of factorization steps: beta = 3 -> beta = 450Improved
increase sample size from N=50 to N=1000 at same betaDistributional Max-Ent Progressive Factorization · GridWorld Block-MDPnumber of factorization steps: N = 50 -> N = 1000Hurt
oracle factorization versus full singleton factorizationDistributional Max-Ent Policy Evaluation · GridWorld Block-MDPKL divergence to true return distributionNo clear effect
full singleton factorization versus oracle factorizationDistributional Max-Ent Policy Evaluation · GridWorld Block-MDPtotal bound BtotHurt
run progressive factorization beyond first splitDistributional Max-Ent Progressive Factorization · GridWorld Block-MDPKL divergenceNo clear effect
minimize the proxy objective J instead of no factorizationDistributional Max-Ent Progressive Factorization · GridWorld Block-MDPKL divergenceImproved

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