AI Paper Atlas

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

SHAP-Guided Kernel Actor-Critic for Explainable Reinforcement Learning

Na Li, Hangguan Shan, Wei Ni, Wenjie Zhang, Xinyu Li

ICML 2026 · Reinforcement learning › Policy gradient and actor-critic · approach: Actor-critic theory · 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
add learned RKHS-SHAP weighting to the actor-critic (vs Advanced AC without SHAP) (ablation)RSA2C-KME · Pendulum-v1 (noise variance 0)mean return: -719.00 -> -139.83Improved
use learned SHAP weights instead of uniform SHAP weights (ablation)RSA2C-KME · Pendulum-v1 (noise variance 0)mean return: -693.08 -> -139.83Improved
use the advanced SHAP-guided actor-critic instead of classical RKHS-AC (ablation)RSA2C-KME · Pendulum-v1 (noise variance 0)mean return: -899.85 -> -139.83Improved
add RKHS-SHAP weighting (vs Advanced AC) (ablation)RSA2C-KME · BipedalWalker-v3 (noise variance 0)mean return: 263.91 -> 279.18Improved
add RKHS-SHAP weighting (vs Advanced AC) (ablation)RSA2C-KME · Ant-v5 (noise variance 0)mean return: 953.62 -> 962.03Improved
use CME instead of KME for SHAP expectationsRSA2C · Pendulum-v1 (noise variance 0)mean return: -139.83 -> -141.86No clear effect
use CME instead of KME for SHAP expectationsRSA2C · BipedalWalker-v3 (noise variance 0.01)mean return: 272.96 -> 269.38No clear effect
increase state noise variance from 0 to 0.01RSA2C-KME · Pendulum-v1mean return: -139.83 -> -153.41Hurt
increase state noise variance from 0 to 0.01PPO · Ant-v5mean return: 677.48 -> -36.31Hurt
increase state noise variance from 0 to 0.01SAC · BipedalWalker-v3return standard deviation: 2.44 -> 82.09Hurt
use CME instead of KME (compute cost)RSA2C · BipedalWalker-v3FLOPs per iteration: 0.381 GFLOPs -> 1.427 GFLOPsHurt
replace SAC (GPU) with RSA2C-KME (CPU)RSA2C-KME · Ant-v5time-to-solution (s): 10287.5 -> 508.2Improved

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