Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Successor-Predecessor Intrinsic Exploration
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 successor-predecessor exploration to DQN | Atari sparse-reward games · Montezuma's Revenge | evaluation score: 1395.4 -> 1530.0 | Helped |
| Add successor-predecessor exploration to DQN | Atari sparse-reward games · Freeway | evaluation score: 29.4 -> 27.5 | Hurt |
| Add successor-predecessor exploration to DQN | Atari sparse-reward games · Gravitar | evaluation score: 1078.3 -> 1223.0 | Helped |
| Add successor-predecessor exploration to DQN | Atari sparse-reward games · Private Eye | evaluation score: 1447.4 -> 488.2 | Hurt |
| Add successor-predecessor exploration to DQN | Atari sparse-reward games · Solaris | evaluation score: 1890.1 -> 2455.8 | Helped |
| Add successor-predecessor exploration to DQN | Atari sparse-reward games · Venture | evaluation score: 1348.5 -> 1274.0 | Hurt |
| Add retrospective predecessor information to SARSA exploration | SARSA · RiverSwim | cumulative reward: 25,075 -> 2,547,156 | Helped |
| Add retrospective predecessor information to SARSA exploration | SARSA · SixArms | cumulative reward: 376,655 -> 2,199,291 | Helped |
| Replace the SR column norm with the PR column norm | SARSA-SRR · RiverSwim | cumulative reward: 2,547,156 -> 2,857,324 | No clear effect |
| Replace the SR column norm with the PR column norm | SARSA-SRR · SixArms | cumulative reward: 2,199,291 -> 1,845,229 | No clear effect |
| Add prospective information to retrospective-only exploration (ablation) | SARSA-SRR · RiverSwim | cumulative reward: 2,629,947 -> 2,547,156 | No clear effect |
| Add prospective information to retrospective-only exploration (ablation) | SARSA-SRR · SixArms | cumulative reward: 1,902,553 -> 2,199,291 | Helped |
Closest papers by meaning
- Reward-Aware Proto-Representations in Reinforcement Learning (NeurIPS 2025 · similarity 0.89)
- Individual Contributions as Intrinsic Exploration Scaffolds for Multi-agent Reinforcement Learning (ICML 2024 · similarity 0.88)
- DISCOVER: Automated Curricula for Sparse-Reward Reinforcement Learning (NeurIPS 2025 · similarity 0.88)
- Improving Intrinsic Exploration by Creating Stationary Objectives (ICLR 2024 · similarity 0.88)
- Action-Dependent Optimality-Preserving Reward Shaping (ICML 2025 · similarity 0.88)
- Decoupling Exploration and Policy Optimization: Uncertainty Guided Tree Search for Hard Exploration (NeurIPS 2026 · 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.