AI Paper Atlas

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

Learning Anisotropic Value Geometry with Finsler Reinforcement Learning

Jumman Hossain, Nirmalya Roy

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 CVaR objective to PPO (ablation)PPO · MuJoCo aggregate (SlopedHopper-12deg, Walker2d-5deg, HalfCheetah-5deg)CVaR0.1 cost: 1.50 -> 1.20Improved
add CVaR objective to PPO (ablation)PPO · MuJoCo aggregate (SlopedHopper-12deg, Walker2d-5deg, HalfCheetah-5deg)energy: 1.00 -> 1.15Hurt
add Finsler cost shaping to PPO (ablation)PPO · MuJoCo aggregate (SlopedHopper-12deg, Walker2d-5deg, HalfCheetah-5deg)energy: 1.00 -> 0.93Improved
add Finsler cost shaping to PPO (ablation)PPO · MuJoCo aggregate (SlopedHopper-12deg, Walker2d-5deg, HalfCheetah-5deg)CVaR0.1 cost: 1.50 -> 1.25Improved
combine Finsler cost with CVaR (FiRL) versus Finsler alone (ablation)PPO + Finsler · MuJoCo aggregate (SlopedHopper-12deg, Walker2d-5deg, HalfCheetah-5deg)CVaR0.1 cost: 1.25 -> 0.80Improved
combine Finsler cost with CVaR (FiRL) versus Finsler alone (ablation)PPO + Finsler · MuJoCo aggregate (SlopedHopper-12deg, Walker2d-5deg, HalfCheetah-5deg)success rate: 94.6 -> 97.4Improved
replace Finsler cost with non-convex ANF cost (ablation)FiRL · RoughTerrainsuccess rate: 95.2% -> 80.4%Hurt
replace Finsler cost with non-convex ANF cost (ablation)FiRL · RoughTerrainenergy (CoT): 0.78 -> 0.95Hurt
use FiRL instead of PPO on 12 degree slopePPO · Hopper, slope 12 degsuccess rate: 82% -> about 98%Improved
lower risk level alpha below 0.05 (ablation)FiRL · SlopedHopper-12degenergy costHurt
add Gaussian angular noise (sigma=5 deg) to terrain-normal estimate (ablation)FiRL · MuJoCo noisy terrainsuccess rate: remains above 94%No clear effect
use FiRL instead of CVaR-PPOCVaR-PPO · Isaac Sim Staircaseenergy: 1.12 -> 0.94Improved

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