Research map / Reinforcement learning / Visual and representation RL
Learning Multi-Level Hierarchies with Hindsight
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 a 2-level hierarchy instead of a flat policy | HAC · All evaluated tasks | episode success rate | Helped |
| Use a 3-level hierarchy instead of a 2-level hierarchy | HAC · All evaluated tasks | episode success rate | Helped |
| Use HAC instead of HIRO | 2-level hierarchical agents · Inverted pendulum, UR5 reacher, ant reacher | success rate | Helped |
| Remove subgoal testing (ablation) | 3-level HAC · Subgoal testing ablation tasks | performance | Hurt |
| Penalize every missed subgoal, including attempts by noisy lower-level policies (ablation) | 3-level HAC · Subgoal testing ablation tasks | performance | Hurt |
Closest papers by meaning
- Data-Efficient Hierarchical Reinforcement Learning (NeurIPS 2018 · similarity 0.91)
- On Efficiency in Hierarchical Reinforcement Learning (NeurIPS 2020 · similarity 0.90)
- Sub-policy Adaptation for Hierarchical Reinforcement Learning (ICLR 2020 · similarity 0.90)
- Hierarchical Decision Making with Structured Policies: A Principled Design via Inverse Optimization (ICML 2026 · similarity 0.90)
- Diversity-Driven Extensible Hierarchical Reinforcement Learning (AAAI 2019 · similarity 0.89)
- Hierarchical Reinforcement Learning with Targeted Causal Interventions (ICML 2025 · similarity 0.88)
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.