Research map / Reinforcement learning / Visual and representation RL
When Waiting Is Not an Option: Learning Options With a Deliberation Cost
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 deliberation cost (λ=0, η=0.020) (ablation) | A2OC · Amidar | final performance: 512.0 -> 880.0 | Helped |
| Add deliberation cost (λ=0, η=0.020) (ablation) | A2OC · Asterix | final performance: 1950.0 -> 5400.0 | Helped |
| Add deliberation cost (λ=0, η=0.020) (ablation) | A2OC · Breakout | final performance: 395.0 -> 430.0 | Helped |
| Add deliberation cost (λ=0, η=0.020) (ablation) | A2OC · Hero | final performance: 2625.0 -> 20100.0 | Helped |
| Add deliberation cost (ablation) | A2OC · Amidar | option termination probability | Helped |
| Increase deliberation cost η (ablation) | A2OC · Atari games | average termination probability | Helped |
| Use deliberation cost with λ=γ, η=0.025 instead of no deliberation cost (ablation) | A2OC · Breakout | final performance: 395.0 -> 414.0 | Helped |
| Use deliberation cost with λ=γ, η=0.025 instead of no deliberation cost (ablation) | A2OC · Hero | final performance: 2625.0 -> 13630.0 | Helped |
| Use deliberation cost with λ=0, η=0.020 instead of no deliberation cost (ablation) | A2OC · Asterix | final performance: 1950.0 -> 5400.0 | Helped |
Closest papers by meaning
- The Option-Critic Architecture (AAAI 2017 · similarity 0.89)
- Reward-Respecting Subtasks for Model-Based Reinforcement Learning (Abstract Reprint) (AAAI 2024 · similarity 0.87)
- No Coin Left Behind: Maximizing Strategic Surplus Against No-Regret Dynamics (NeurIPS 2026 · similarity 0.86)
- Autonomous Option Invention for Continual Hierarchical Reinforcement Learning and Planning (AAAI 2025 · similarity 0.86)
- Exploiting Action Impact Regularity and Exogenous State Variables for Offline Reinforcement Learning (Abstract Reprint) (AAAI 2024 · similarity 0.85)
- Optimism Without Regularization: Constant Regret in Zero-Sum Games (NeurIPS 2025 · similarity 0.85)
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.