Research map / Reinforcement learning / Visual and representation RL
Hierarchical Reinforcement Learning via Advantage-Weighted Information Maximization
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 adInfoHRL instead of TD3 | TD3 · Walker2d | return | Helped |
| use adInfoHRL instead of TD3 | TD3 · Ant-rllab | return | Helped |
| use adInfoHRL instead of TD3 | TD3 · HalfCheetah | return | No clear effect |
| use adInfoHRL instead of TD3 | TD3 · Hopper | return | Mixed |
| add advantage-weighted importance when computing mutual information (ablation) | infoHRL · continuous control tasks | return | Helped |
| use two options instead of four (ablation) | adInfoHRL · Walker2d-v1 | return: 3752.1 $\pm$ 442 vs 3404.2$\pm$ 785.6 | Helped |
| use two options instead of four (ablation) | adInfoHRL · HalfCheetah-v1 | return: 6315.1 $\pm$ 612.8 vs 4520.6 $\pm$ 859.3 | Helped |
| use two options instead of four (ablation) | adInfoHRL · Hopper-v1 | return: 1821.7 $\pm$ 626.3 vs 1717.5 $\pm$ 281.7 | No clear effect |
| use two options instead of four (ablation) | adInfoHRL · Ant-rllab | return: 1263.2 $\pm$ 333.5 vs 683.2 $\pm$ 105.68 | Helped |
| use adInfoHRL instead of IOPG | IOPG · Walker2d-v1, HalfCheetah-v1, Hopper-v1 | return | Helped |
| use adInfoHRL instead of SAC-LSP | SAC-LSP · Walker2d-v1 and Ant-rllab | return | Helped |
| use adInfoHRL instead of SAC-LSP | SAC-LSP · HalfCheetah-v1 and Hopper-v1 | return | Hurt |
Closest papers by meaning
- Hierarchical Reinforcement Learning with Advantage-Based Auxiliary Rewards (NeurIPS 2019 · similarity 0.92)
- On Efficiency in Hierarchical Reinforcement Learning (NeurIPS 2020 · similarity 0.92)
- Data-Efficient Hierarchical Reinforcement Learning (NeurIPS 2018 · similarity 0.91)
- Deep Reinforcement Learning from Hierarchical Preference Design (ICML 2025 · similarity 0.90)
- Hierarchical Reinforcement Learning with Uncertainty-Guided Diffusional Subgoals (ICML 2025 · similarity 0.89)
- Probabilistic Subgoal Representations for Hierarchical Reinforcement Learning (ICML 2024 · 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.