AI Paper Atlas

Research map / Reinforcement learning / Offline RL

RAMBO-RL: Robust Adversarial Model-Based Offline Reinforcement Learning

Marc Rigter, Bruno Lacerda, Nick Hawes

NeurIPS 2022 · Reinforcement learning › Offline RL · approach: Simple practical offline RL · 18 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 adversarial model updates to RAMBO (ablation)RAMBO · MuJoCo-v2normalized return: 694.4 ± 56.5 -> 826.2 ± 33.8Improved
Add adversarial model updates to RAMBO (ablation)RAMBO · AntMaze-v0normalized return: 45.8 ± 21.8 -> 67.0 ± 14.9Improved
Choose hyperparameters offline using the Q-value magnitude and stability heuristic instead of online tuningRAMBO · MuJoCo-v2normalized return: 826.2 ± 33.8 -> 785.5 ± 40.4Hurt
Choose hyperparameters offline using the Q-value magnitude and stability heuristic instead of online tuningRAMBO · AntMaze-v0normalized return: 67.0 ± 14.9 -> 37.8 ± 12.4Hurt
Use adversarial model training instead of COMBO's value-function pessimismRAMBO vs COMBO · Single Transition Exampleperformance: 1.41 ± 0.12 -> 1.49 ± 0.05Improved
Use adversarial weighting λ = 0 instead of λ = 3e-4 (ablation)RAMBO · MuJoCo-v2 Random HalfCheetahnormalized return: 34.5 ± 3.7 -> 40.0 ± 2.3Mixed
Use adversarial weighting λ = 0 instead of λ = 3e-4 (ablation)RAMBO · MuJoCo-v2 Medium HalfCheetahnormalized return: 77.6 ± 1.5 -> 75.5 ± 6.0Hurt
Use rollout length k = 5 instead of k = 2, with λ = 3e-4 (ablation)RAMBO · MuJoCo-v2 Medium HalfCheetahnormalized return: 72.5 ± 4.4 -> 77.6 ± 1.5Improved
Use rollout length k = 5 instead of k = 2, with λ = 3e-4 (ablation)RAMBO · MuJoCo-v2 Random Walker2Dnormalized return: − -> 0.2 ± 0.5Mixed
Use behavior-cloning policy initialization (ablation)RAMBO · MuJoCo-v2normalized returnImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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