Research map / Reinforcement learning / Offline RL
Supported Policy Optimization for Offline Reinforcement Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| replace VAE density estimator with tanh-Gaussian behavior model (ablation) | SPOT · AntMaze-medium-play-v2 | normalized score: 74.7 -> 46.0 | Hurt |
| replace VAE density estimator with tanh-Gaussian behavior model (ablation) | SPOT · Hopper-medium-expert-v2 | normalized score: 99.3 -> 79.9 | Hurt |
| replace VAE density estimator with tanh-Gaussian behavior model (ablation) | SPOT · Hopper-medium-v2 | normalized score: 86.0 -> 96.6 | Helped |
| remove Q normalization (ablation) | SPOT · Gym-MuJoCo and AntMaze | normalized score: 86.0 -> 101.2 (Hopper-m); 74.7 -> 72.2 (AntMaze-m-p) | Mixed |
| use SAC base with stochastic policy and no entropy instead of TD3 (ablation) | SPOT · AntMaze-medium-diverse-v2 | normalized score: 79.1 -> 0.0 | Hurt |
| use SAC base with entropy regularization instead of TD3 (ablation) | SPOT · AntMaze-medium-diverse-v2 | normalized score: 79.1 -> 22.2 | Hurt |
| use SAC-style base instead of TD3 (ablation) | SPOT · Gym-MuJoCo | normalized score: 86.0 -> 58.4 (Hopper-m, w/o Ent) | Mixed |
| increase VAE importance samples L from 1 to 5 or 10 (ablation) | SPOT · Gym-MuJoCo | normalized score | No clear effect |
| SPOT vs IQL on offline AntMaze | SPOT · AntMaze total | normalized score total: 392.4 -> 359.6 | Hurt |
| online fine-tune from SPOT offline init with lambda decayed to one-fifth | SPOT · AntMaze online | normalized score total: 367.8 -> 569.0 | Helped |
Closest papers by meaning
- Iteratively Refined Behavior Regularization for Offline Reinforcement Learning (NeurIPS 2024 · similarity 0.91)
- Adaptive Advantage-Guided Policy Regularization for Offline Reinforcement Learning (ICML 2024 · similarity 0.91)
- Behavior-Regularized Diffusion Policy Optimization for Offline Reinforcement Learning (ICML 2025 · similarity 0.90)
- A Clean Slate for Offline Reinforcement Learning (NeurIPS 2025 · similarity 0.89)
- Constrained Latent Action Policies for Model-Based Offline Reinforcement Learning (NeurIPS 2024 · similarity 0.89)
- Relaxed Stationary Distribution Correction Estimation for Improved Offline Policy Optimization (AAAI 2024 · similarity 0.89)
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.