Research map / Reinforcement learning / Visual and representation RL
Successor Features for Transfer in 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 |
|---|---|---|---|
| Use SFQL with successor features and generalized policy improvement instead of PRQL | Four-room navigation · Four-room navigation tasks | average return: more than 100% | Helped |
| Use PRQL instead of QL | Four-room navigation · Four-room navigation tasks | average return: around 100% | Helped |
| Learn reward-predicting features from data rather than use features that perfectly predict rewards (ablation) | SFQL · Four-room navigation tasks | average return | Mixed |
| Train on one task while sharing successor features across tasks | SFDQN · Reacher domain, including test tasks | normalized return | Helped |
Closest papers by meaning
- Reward-Aware Proto-Representations in Reinforcement Learning (NeurIPS 2025 · similarity 0.88)
- Distributional Successor Features Enable Zero-Shot Policy Optimization (NeurIPS 2024 · similarity 0.88)
- Learning Generalizable and Composable Abstractions for Transfer in Reinforcement Learning (AAAI 2024 · similarity 0.87)
- SF-DQN: Provable Knowledge Transfer using Successor Feature for Deep Reinforcement Learning (ICML 2024 · similarity 0.87)
- Proto Successor Measure: Representing the Behavior Space of an RL Agent (ICML 2025 · similarity 0.86)
- How Ensembles of Distilled Policies Improve Generalisation in Reinforcement Learning (NeurIPS 2025 · similarity 0.86)
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.