AI Paper Atlas

Research map / Reinforcement learning / MDPs and planning

ODE-based Recurrent Model-free Reinforcement Learning for POMDPs

Xuanle Zhao, Duzhen Zhang, Liyuan Han, Tielin Zhang, Bo XU

NeurIPS 2023 · Reinforcement learning › MDPs and planning · approach: POMDPs and confounded RL · 2 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
separate context encoders for policy and value instead of shared (ablation)PODMP · Walker-Pfinal return: 1336 -> 1487Improved
separate context encoders for policy and value instead of shared (ablation)PODMP · Walker-Vfinal return: 584 -> 612Improved
context encoder input o,a,r instead of o only (ablation)PODMP · Walker-Pfinal return: 1276 -> 1487Improved
add previous action to observation input (oa) vs o (ablation)PODMP · Walker-Pfinal return: 1276 -> 1065Hurt
add reward to observation and action input (oar vs oa) (ablation)PODMP · Walker-Pfinal return: 1065 -> 1487Improved
RK4 solver instead of Euler (ablation)PODMP · Ant-Pfinal return: 1243 -> 1183No clear effect
RK4 solver instead of Euler (ablation)PODMP · Walker-Vfinal return: 612 -> 683No clear effect
Heun solver instead of Euler (ablation)PODMP · Ant-Pfinal return: 1243 -> 1123No clear effect
RK4 solver instead of Euler (ablation)PODMP · AntBLT-P training timehours: 12 hours -> 20 hoursHurt
Heun solver instead of Euler (ablation)PODMP · AntBLT-P training timehours: 12 hours -> 17 hoursHurt
GRU-ODE encoder (ours) vs RMF baselinePODMP · Hopper-Pfinal return: 2133 -> 2455Improved
GRU-ODE encoder (ours) vs RMF baselinePODMP · Ant-Vfinal return: 1021 -> 998Hurt

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