Research map / Reinforcement learning / Policy gradient and actor-critic
The Option-Critic Architecture
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 |
|---|---|---|---|
| Learn intra-option policies and termination conditions jointly with the policy over options | Option-Critic · Four-rooms | recovery speed | Helped |
| Learn options end-to-end with intra-option Q-learning | Option-Critic · Pinball | learning speed | Helped |
| Learn options end-to-end using deep function approximation | Option-Critic · Asterisk | performance | Helped |
| Learn options end-to-end using deep function approximation | Option-Critic · Seaquest | performance | Helped |
| Learn options end-to-end using deep function approximation | Option-Critic · Zaxxon | performance | Helped |
| Add a baseline to the intra-option policy-gradient estimator (ablation) | Option-Critic · Seaquest | intra-option policy quality | Helped |
| Add a baseline to the intra-option policy-gradient estimator (ablation) | Option-Critic · Arcade Learning Environment | agent performance | Helped |
| Add a termination-gradient margin regularizer (ablation) | Option-Critic · Arcade Learning Environment | option duration | Helped |
| Add entropy regularization to intra-option policies (ablation) | Option-Critic · Arcade Learning Environment | policy entropy | Helped |
Closest papers by meaning
- Flexible Option Learning (NeurIPS 2021 · similarity 0.93)
- On the Role of Weight Sharing During Deep Option Learning (AAAI 2020 · similarity 0.90)
- Options of Interest: Temporal Abstraction with Interest Functions (AAAI 2020 · similarity 0.90)
- Autonomous Option Invention for Continual Hierarchical Reinforcement Learning and Planning (AAAI 2025 · similarity 0.90)
- Reward-Respecting Subtasks for Model-Based Reinforcement Learning (Abstract Reprint) (AAAI 2024 · similarity 0.90)
- Option-aware Temporally Abstracted Value for Offline Goal-Conditioned Reinforcement Learning (NeurIPS 2025 · similarity 0.87)
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.