Research map / Multi-agent systems and games / Cooperative multi-agent RL
Delayed Propagation Transformer: A Universal Computation Engine towards Practical Control in Cyber-Physical Systems
Wenqing Zheng, Qiangqiang Guo, Hao Frank Yang, Peihao Wang, Zhangyang Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| DePT instead of Co-light teacher | traffic signal control · Grid-Bi | AvgTT: 191.66 -> 184.22 | Improved |
| DePT instead of Co-light | traffic signal control · Grid-Uni | AvgTT: 188.23 -> 180.71 | Improved |
| DePT instead of Co-light | traffic signal control · NewYork | AvgTT: 1605.56 -> 1555.47 | Improved |
| DePT instead of Co-light | traffic signal control · NewYork | AvgQue: 1.69 -> 1.52 | Improved |
| DePT instead of Co-light | traffic signal control · Grid-Bi | AvgQue: 0.05 -> 0.05 | No clear effect |
| DePT instead of Max-pressure | traffic signal control · Grid-Bi | AvgTT: 208.13 -> 184.22 | Improved |
| remove prior pre-fitting (ablation) | DePT · Grid-Bi | AvgTT (round 30): 244.3 -> 782.1 | Hurt |
| remove prior pre-fitting (ablation) | DePT · Grid-Bi | AvgTT (round 190): 184.9 -> 189.2 | Hurt |
| remove prior pre-fitting (ablation) | DePT · Grid-Bi | AvgTT (round 0): 1143.1 -> 997.2 | Improved |
| remove ConeDecay prior (ablation) | DePT · Grid-Bi | AvgTT (round 190): 184.9 -> 269.3 | Hurt |
| remove all priors (traditional transformer encoder) (ablation) | DePT · Grid-Bi | AvgTT (round 190): 184.9 -> 313.7 | Hurt |
| remove ConeDecay prior (ablation) | DePT · Grid-Bi | AvgTT (round 130 vs 100): 188.2 -> 240.3 | Hurt |
Closest papers by meaning
- BehaviorGPT: Smart Agent Simulation for Autonomous Driving with Next-Patch Prediction (NeurIPS 2024 · similarity 0.81)
- Towards Efficient Collaboration via Graph Modeling in Reinforcement Learning (AAAI 2025 · similarity 0.81)
- Bridging Training and Execution via Dynamic Directed Graph-Based Communication in Cooperative Multi-Agent Systems (AAAI 2025 · similarity 0.81)
- CoLLMLight: Cooperative Large Language Model Agents for Network-Wide Traffic Signal Control (ICLR 2026 · similarity 0.80)
- Augmented Runtime Collaboration for Self-Organizing Multi-Agent Systems: A Hybrid Bi-Criteria Routing Approach (AAAI 2026 · similarity 0.80)
- Do Transformer World Models Give Better Policy Gradients? (ICML 2024 · similarity 0.80)
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.
