Research map / Reinforcement learning / Offline RL
Instabilities of Offline RL with Pre-Trained Neural Representation
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 |
|---|---|---|---|
| Add random-policy trajectories to the target-policy dataset | LSTD with pre-trained neural network features · Hopper-v2 and Walker2d-v2 | RMSE: Hopper-v2: $2.18 \pm 1.14$ -> $9.38 \pm 3.84$ (0.5x), $13.18 \pm 2.77$ (1x), $16.86 \pm 2.84$ (2x); Walker2d-v2: $13.88 | Hurt |
| Add random-policy trajectories to the target-policy dataset | FQI with pre-trained neural network features · OpenAI Gym tasks | RMSE | Hurt |
| Add data from lower-performance policies to the target-policy dataset | FQI with pre-trained neural network features · OpenAI Gym tasks | RMSE | Mixed |
| Use random Fourier features instead of pre-trained neural network features | FQI · OpenAI Gym tasks with random-policy data | RMSE | Hurt |
| Vary the regularization parameter λ while adding random-policy data | FQI with pre-trained neural network features · Walker2d-v2 | RMSE | Mixed |
Closest papers by meaning
- Efficient Online Reinforcement Learning Fine-Tuning Need Not Retain Offline Data (ICLR 2025 · similarity 0.90)
- Is Value Learning Really the Main Bottleneck in Offline RL? (NeurIPS 2024 · similarity 0.90)
- Tackling Data Corruption in Offline Reinforcement Learning via Sequence Modeling (ICLR 2025 · similarity 0.90)
- Neural Network Approximation for Pessimistic Offline Reinforcement Learning (AAAI 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 0.90)
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.