Research map / Reinforcement learning / Offline RL
PerSim: Data-Efficient Offline Reinforcement Learning with Heterogeneous Agents via Personalized Simulators
Anish Agarwal, Abdullah Omar Alomar, Varkey Alumootil, Devavrat Shah, Dennis Shen, Zhi Xu, Cindy Yang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Augment BCQ training data with 5 PerSim-generated trajectories per agent | BCQ · CartPole (2/0.5) | average reward: 43.90 -> 80.04 | Improved |
| Augment CQL training data with 5 PerSim-generated trajectories per agent | CQL · CartPole (10.0/0.5) | average reward: 9.300 -> 198.3 | Improved |
| Augment BCQ with PerSim trajectories | BCQ · MountainCar 0.01 | average reward: -500.0 -> -68.84 | Improved |
| Augment BCQ/CQL with PerSim trajectories | BCQ/CQL · MountainCar 0.035 | average reward: -500.0 -> -500.0 | No clear effect |
| Augment CQL with PerSim trajectories | CQL · HalfCheetah (0.3/0.3) | average reward: -0.600 -> -402.2 | Hurt |
| Augment BCQ with PerSim trajectories | BCQ · HalfCheetah (0.3/1.7) | average reward: -498.9 -> 571.8 | Improved |
| Use PerSim instead of PE-TS CaDM with MPC | CaDM · MountainCar 0.025 reward | average reward: -492.3 -> -189.4 | Improved |
| Use PerSim instead of BCQ-P | BCQ-P · HalfCheetah (1.7/0.3) reward | average reward: 2006 -> 997.0 | Hurt |
| Reduce number of training agents from N=250 to N=25 (ablation) | PerSim · reward under data scarcity | average reward | No clear effect |
Closest papers by meaning
- Learning from Reward-Free Offline Data: A Case for Planning with Latent Dynamics Models (NeurIPS 2025 · similarity 0.90)
- Constrained Latent Action Policies for Model-Based Offline Reinforcement Learning (NeurIPS 2024 · similarity 0.90)
- A Clean Slate for Offline Reinforcement Learning (NeurIPS 2025 · similarity 0.90)
- Forecasting in Offline Reinforcement Learning for Non-stationary Environments (NeurIPS 2025 · similarity 0.89)
- Beyond OOD State Actions: Supported Cross-Domain Offline Reinforcement Learning (AAAI 2024 · similarity 0.89)
- Simple Ingredients for Offline Reinforcement Learning (ICML 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.
