AI Paper Atlas

Research map / Reinforcement learning / Offline RL

One Risk to Rule Them All: A Risk-Sensitive Perspective on Model-Based Offline Reinforcement Learning

Marc Rigter, Bruno Lacerda, Nick Hawes

NeurIPS 2023 · Reinforcement learning › Offline RL · approach: Simple practical offline 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
remove risk-averse sampling (ablation)1R2R_CVaR · Currency Exchangestatic CVaR0.1: 31.4 -> 64.0Improved
remove risk-averse sampling (ablation)1R2R_CVaR · HIV Treatmentstatic CVaR0.1: 41.2 -> 53.9Improved
remove risk-averse sampling (ablation)1R2R_CVaR · D4RL Medium-Replay Hoppernormalised expected value: 55.3 -> 92.9Improved
remove risk-averse sampling (ablation)1R2R_CVaR · D4RL Random Hoppernormalised expected value: div. -> 30.9Improved
remove risk-averse sampling (ablation)1R2R_CVaR · Stochastic MuJoCo Medium-Replay Hopper (High)normalised static CVaR0.1: 30.5 -> 51.4Improved
remove risk-averse sampling (ablation)1R2R_CVaR · Currency Exchangeaverage return: 99.7 vs 78.8Hurt
remove model ensemble (ablation)1R2R_CVaR · D4RL Medium-Expert Hoppernormalised expected value: 45.8 -> 81.6Improved
remove model ensemble (ablation)1R2R_CVaR · D4RL Medium HalfCheetahnormalised expected value: 60.3 -> 74.5Improved
use Wang risk measure instead of CVaR1R2R · Stochastic MuJoCo Medium-Replay Walker2D (High)normalised static CVaR0.1: 46.0 -> 28.8Hurt
use larger ensemble of 20 networks (ablation)1R2R_CVaR · Stochastic MuJoCo Medium-Replay Walker2D (High)normalised static CVaR0.1: 46.0 -> 39.1Hurt
sample 50 successor states instead of fewer (ablation)1R2R_CVaR · Stochastic MuJoCo Medium-Replay Walker2D (High)normalised static CVaR0.1: 46.0 -> 39.6Hurt
train entirely on synthetic data (f=0) (ablation)1R2R_CVaR · Stochastic MuJoCo Medium-Replay Walker2D (High)normalised static CVaR0.1: 46.0 -> 1.8Hurt

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