Research map / Reinforcement learning / Visual and representation RL
Hierarchy Through Composition with Multitask LMDPs
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 a hierarchy with subtasks every log N states | Multitask LMDP · 1D ring, N tasks | total iterations: O(N 2) -> O(N log N) | Helped |
| Add hierarchy to memory encoding | Multitask LMDP · 1D ring, N tasks | nonzero elements of Z: O(N 2) -> O(N log N) | Helped |
| Add omniscient hierarchy to base-layer Z-learning (ablation) | Base layer policy · Rooms domain | trajectory length | Helped |
| Hierarchy advantage applies to single task | Flat scheme · Ring, one specific task | iterations | No clear effect |
| Use composed multitask LMDP agent instead of options Q-learning | Office building OR task · Office building navigation | trajectory length over epochs | Helped |
Closest papers by meaning
- Globally Optimal Hierarchical Reinforcement Learning for Linearly-Solvable Markov Decision Processes (AAAI 2022 · similarity 0.89)
- Hierarchical Subtask Discovery with Non-Negative Matrix Factorization (ICLR 2018 · similarity 0.88)
- Learning Multi-Level Hierarchies with Hindsight (ICLR 2019 · similarity 0.87)
- An Efficient Approach to Model-Based Hierarchical Reinforcement Learning (AAAI 2017 · similarity 0.87)
- Robust Subtask Learning for Compositional Generalization (ICML 2023 · similarity 0.87)
- Representation Learning Enables Scalable Multitask Deep Reinforcement Learning (NeurIPS 2026 · 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.