Research map / Reinforcement learning / Visual and representation RL
Graying the black box: Understanding DQNs
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 |
|---|---|---|---|
| Eject trajectories predicted likely to be low-reward using the SAMDP model | DQN on Breakout · Breakout | average trajectory reward: 36% | Helped |
| Eject trajectories predicted likely to be low-reward using the SAMDP model | DQN on Seaquest · Seaquest | average trajectory reward: 20% | Helped |
| Eject trajectories predicted likely to be low-reward using the SAMDP model | DQN on Pacman · Pacman | average trajectory reward: 4.7% | Helped |
| Select SAMDP by intersecting top-ranked models over four criteria instead of using random clusters (ablation) | SAMDP model selection · Atari games | all four evaluation criteria: none of 10000 random models scored better | Helped |
| Greedy SAMDP policy compared against top vs bottom rewarded trajectories | SAMDP · Atari games | correlation with transition matrices | Helped |
Closest papers by meaning
- Shallow Updates for Deep Reinforcement Learning (NeurIPS 2017 · similarity 0.90)
- Convergent and Efficient Deep Q Learning Algorithm (ICLR 2022 · similarity 0.89)
- Deep Reinforcement Learning with Double Q-Learning (AAAI 2016 · similarity 0.89)
- Confounding Robust Deep Reinforcement Learning: A Causal Approach (NeurIPS 2025 · similarity 0.88)
- Beyond The Rainbow: High Performance Deep Reinforcement Learning on a Desktop PC (ICML 2025 · similarity 0.86)
- Deep Double Q-learning (NeurIPS 2026 · 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.