AI Paper Atlas

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

Geometric Coherence Learning for Structuring Value Functions in Plain MDPs

Zuyuan Zhang, Zeyu Fang, Tian Lan

ICML 2026 · Reinforcement learning › Policy gradient and actor-critic · approach: Temporal-difference and discounting theory

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
add logic-order regularization to symmetry-only (ablation)GCR-RL · E4 Noisy-RPS Chainfinal return: 20.7 ± 0.01 -> 21.60 ± 0.20Improved
add symmetry (equivariance) losses to logic-only (ablation)GCR-RL · E4 Noisy-RPS Chainfinal return: 20.9 ± 0.00 -> 21.60 ± 0.20Improved
add learned action relabeling (ablation)GCR-RL · E2 ButtonPermutationfinal return: 2.010 ± 0.01 -> 3.700 ± 0.00Improved
add learned action relabeling (ablation)GCR-RL · E4 Noisy-RPS Chainfinal return: 19.70 ± 0.21 -> 21.60 ± 0.20Improved
add logic-order to symmetry-only (ablation)GCR-RL · E1 RotMirror-Gridfinal return: −144.3 ± 5.2 -> −136.7 ± 7.1Improved
add symmetry losses to logic-only (ablation)GCR-RL · E1 RotMirror-Gridfinal return: −153.2 ± 10.1 -> −136.7 ± 7.1Improved
add logic-order to symmetry-only (ablation)GCR-RL · Cartpolefinal return: 292.6 ± 32.1 -> 321.9 ± 52.7Improved
add action relabeling (ablation)GCR-RL · Cartpolefinal return: 314.5 ± 13.2 -> 321.9 ± 52.7No clear effect
use symmetry regularizer weight λeq=0.10 instead of 0 (ablation)GCR-RL · E4 Noisy-RPS Chainfinal reward at 200 epochs: 16.2 ± 1.8 -> 18.7 ± 0.8Improved
use order weight λiso=0.10 instead of 0 (ablation)GCR-RL · E4 Noisy-RPS Chainfinal reward at 200 epochs: 15.4 ± 2.0 -> 18.7 ± 0.8Improved
raise λeq from 0.10 to 0.40 (ablation)GCR-RL · E4 Noisy-RPS Chainfinal reward at 200 epochs: 18.7 ± 0.8 -> 17.1 ± 1.5Hurt
use K=4 transforms instead of 1 (ablation)GCR-RL · E4 Noisy-RPS Chainfinal reward at 200 epochs: 16.6 ± 1.6 -> 18.7 ± 0.8Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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