AI Paper Atlas

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

Frictional Q-Learning

Hyunwoo Kim, Hyo Kyung Lee

ICML 2026 · Reinforcement learning › Policy gradient and actor-critic · approach: Regularized scalable deep 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
Add total correlation term to the cVAE (ablation)FQL · Ant-v4average return: 3204.6 -> 5102.1Improved
Add total correlation term to the cVAE (ablation)FQL · Hopper-v4average return: 3370.5 -> 3370.5No clear effect
Use latent dimension 2 dim(A) instead of 1 (ablation)FQL · Walker2d-v4average return: 4201.0 -> 4600.2Improved
Use latent dimension 2 dim(A) instead of 1 (ablation)FQL · HalfCheetah-v4average return: 10096.0 -> 11429.0Improved
Use argmin Q background selection instead of uniform sampling (ablation)FQL · Ant-v4average return: 3160.1 -> 5102.1Improved
Use argmin Q background selection instead of uniform sampling (ablation)FQL · Walker2d-v4average return: 4729.1 -> 4600.2Hurt
Use replay buffer 1M instead of 0.1M (ablation)FQL · MuJoCo (5 envs)average return: Walker2d 2965.0 -> 4600.2; HalfCheetah 11939.1 -> 11429.0Mixed
Vary beta from 0 to 2 (ablation)FQL · Hopper-v4average return: 2973.0 -> 3370.5Improved
Vary beta from 0 to 1 (ablation)FQL · HalfCheetah-v4average return: 13497.0 -> 15969.6Improved
Use FQL instead of SACSAC · Humanoid-v4average return: 5682.62 -> 6437.15Improved
Use FQL instead of TD3TD3 · Hopper-v4average return: 3371.87 -> 3370.51No clear effect
Use FQL instead of TD3TD3 · Ant-v4average return: 6542.34 -> 6176.60No clear effect

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