Research map / Reinforcement learning / Visual and representation RL
Using Reward Machines for High-Level Task Specification and Decomposition 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 |
|---|---|---|---|
| Decompose reward machine into per-state subpolicies learned with off-policy Q-learning (QRM) vs. standard Q-learning on each task | QRM · Office World | normalized discounted reward | Helped |
| Replace Q-learning with deep Double DQN and prioritized replay in reward machine decomposition (DQRM) | DQRM · Water World | normalized discounted reward | Helped |
| Use options-based hierarchical RL instead of reward-machine decomposition | HRL · Minecraft World | normalized discounted reward | Hurt |
| Add reward machine knowledge to HRL by pruning options (HRL-RM) | HRL-RM · Office World | normalized discounted reward | No clear effect |
Closest papers by meaning
- Discriminative Policy Optimization for Token-Level Reward Models (ICML 2025 · similarity 0.90)
- Learning Reward Machines for Partially Observable Reinforcement Learning (NeurIPS 2019 · similarity 0.90)
- Hierarchies of Reward Machines (ICML 2023 · similarity 0.89)
- Inverting the Bellman Equation: From $Q$-Values to World Models (NeurIPS 2026 · similarity 0.87)
- Efficient Reinforcement Learning in Probabilistic Reward Machines (AAAI 2025 · similarity 0.87)
- Switching the Loss Reduces the Cost in Batch Reinforcement Learning (ICML 2024 · 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.