Research map / Multi-agent systems and games / Cooperative multi-agent RL
Mingling Foresight with Imagination: Model-Based Cooperative Multi-Agent Reinforcement Learning
Zhiwei Xu, Dapeng Li, Bin Zhang, Yuan Zhan, Yunpeng Baiia, Guoliang Fan
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 |
|---|---|---|---|
| Add latent imagined rollouts to QMIX (MBVD) | QMIX · SMAC | sample efficiency | Improved |
| Add latent imagined rollouts to QMIX (MBVD) | QMIX · Google Research Football | sample efficiency | Improved |
| Add latent imagined rollouts to QMIX (MBVD) | QMIX · Multi-Agent Discrete MuJoCo | episode return | Improved |
| Replace imagined rollouts with aggregated real states (QMIX-RS) (ablation) | QMIX · SMAC 2c_vs_64zg | task performance | Hurt |
| Replace imagined rollouts with aggregated real latent states (QMIX-LS) (ablation) | QMIX · SMAC 2c_vs_64zg | task performance | Hurt |
| Use latent states rather than real states for the aggregated next-step information (ablation) | QMIX-LS vs. QMIX-RS · SMAC 2c_vs_64zg | task performance | Improved |
| Use short imagined rollout horizons (ablation) | MBVD · SMAC 2c_vs_64zg | task performance | No clear effect |
| Increase the imagined rollout horizon substantially (ablation) | MBVD · SMAC 2c_vs_64zg | task performance | Hurt |
| Use shorter rollout horizons (ablation) | MBVD · SMAC 1c3s5z | sample efficiency | Improved |
| Use longer rollout horizons (ablation) | MBVD · SMAC MMM2 | task performance | Improved |
| Use a fixed rollout horizon across tasks (ablation) | MBVD · SMAC | task performance | Mixed |
Closest papers by meaning
- Acting Beyond Learning: Imagination-Assisted Decision-Making in the Visual-based Multi-Agent Cooperative Scenarios (AAAI 2025 · similarity 0.88)
- Models as Agents: Optimizing Multi-Step Predictions of Interactive Local Models in Model-Based Multi-Agent Reinforcement Learning (AAAI 2023 · similarity 0.87)
- Imagine, Initialize, and Explore: An Effective Exploration Method in Multi-Agent Reinforcement Learning (AAAI 2024 · similarity 0.86)
- Optimistic Value Instructors for Cooperative Multi-Agent Reinforcement Learning (AAAI 2024 · similarity 0.86)
- Revisiting Multi-Agent World Modeling from a Diffusion-Inspired Perspective (NeurIPS 2025 · similarity 0.85)
- Grounded Answers for Multi-agent Decision-making Problem through Generative World Model (NeurIPS 2024 · 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.
