Research map / Multi-agent systems and games / Cooperative multi-agent RL
Off-Policy Action Anticipation in Multi-Agent Reinforcement Learning
Ariyan Bighashdel, Daan de Geus, Pavol Jancura, Gijs Dubbelman
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 |
|---|---|---|---|
| use action anticipation (LOLA-OffPA2) instead of policy parameter anticipation (LOLA-DiCE) | LOLA · Exit-Room level 3 | normalized average episode reward: 0.56 -> 0.93 | Improved |
| use LOLA-OffPA2 instead of LOLA-DiCE | LOLA · Exit-Room, 4th-order reasoning | learning anticipation time complexity: 5.41 -> 0.94 | Improved |
| use LOLA-OffPA2 instead of LOLA-DiCE | LOLA · Exit-Room level 1 | normalized average episode reward: 0.91 -> 1.00 | Improved |
| use HLA-OffPA2 instead of MADDPG | MADDPG · Reacher (Mujoco) | normalized average episode reward: 0.02 -> 0.42 | Improved |
| use HLA-OffPA2 instead of MADDPG | MADDPG · Predator-Prey | normalized average episode reward: 0.21 -> 0.44 | Improved |
| use HLA-OffPA2 instead of CPG-MADDPG | CPG-MADDPG · Half-Cheetah | normalized average episode reward: 0.88 -> 0.94 | Improved |
| add sorting of hierarchy levels by shaping capacity (ablation) | HLA-OffPA2 · Walker | normalized average episode reward: 0.63 -> 0.67 | Improved |
| add sorting of hierarchy levels by shaping capacity (ablation) | HLA-OffPA2 · Cooperative Navigation | normalized average episode reward: 0.85 -> 0.88 | Improved |
| add sorting of hierarchy levels by shaping capacity (ablation) | HLA-OffPA2 · Predator-Prey | normalized average episode reward: 0.38 -> 0.44 | Improved |
| raise projected prediction length from tuned value to 1.3 (ablation) | LOLA-OffPA2 · Exit-Room level 3 | normalized average episode reward | Hurt |
| lower projected prediction length to 0.1 (ablation) | LOLA-OffPA2 · Exit-Room level 3 | normalized average episode reward | Hurt |
| use LA-OffPA2 instead of MADDPG-based methods | MADDPG · Iterated Rotational Game | distance to equilibrium | Improved |
Closest papers by meaning
- Offline Multi-Agent Reinforcement Learning via In-Sample Sequential Policy Optimization (AAAI 2025 · similarity 0.88)
- ComaDICE: Offline Cooperative Multi-Agent Reinforcement Learning with Stationary Distribution Shift Regularization (ICLR 2025 · similarity 0.87)
- Revisiting Cooperative Off-Policy Multi-Agent Reinforcement Learning (ICML 2025 · similarity 0.86)
- Dispelling the Mirage of Progress in Offline MARL through Standardised Baselines and Evaluation (NeurIPS 2024 · similarity 0.86)
- Efficient Multi-agent Reinforcement Learning by Planning (ICLR 2024 · similarity 0.86)
- Mimicking To Dominate: Imitation Learning Strategies for Success in Multiagent Games (NeurIPS 2024 · similarity 0.86)
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.
