Constrained Multi-Agent Reinforcement Learning in Imagination
Predicting cost in a multi-agent world model moves most constrained policy updates into imagination: the stronger model-free baseline needs $10$ to $20$ times CA-MAMBA's interactions on the median SMAC configuration, and CA-MAMBA earns $2.8$ to $12$ times the baselines' MAMuJoCo reward at equal budget.
ProofPaper ↗
Key points
- In controlled tests on one task per benchmark, pricing the model's cost advantages at parity more than halves a cost-blind learner's cost and meets a tight continuous-control limit the learned multiplier misses, suggesting the model's cost estimates are not the bottleneck.
Sources (1)
- [1]Constrained Multi-Agent Reinforcement Learning in ImaginationarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 15, 06:09 AM
Predicting cost in a multi-agent world model moves most constrained policy updates into imagination: the stronger model-free baseline needs $10$ to $20$ times CA-MAMBA's interactions on the median SMAC configuration, and CA-MAMBA earns $2.8$ to $12$ times the baselines' MAMuJoCo reward at equal budget.
In controlled tests on one task per benchmark, pricing the model's cost advantages at parity more than halves a cost-blind learner's cost and meets a tight continuous-control limit the learned multiplier misses, suggesting the model's cost estimates are not the bottleneck.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 15, 2026Learning to Think and Explore in 3D World
- Oct 15, 2026GlobeBench: Evaluating Text-Based World Model Capabilities for Scalable Environment Simulations
- Oct 14, 2026Sign Betting for Misspecified Regression: Variance Adaptivity and Optimal Approximation
- Oct 14, 2026When Reality Outpaces Cognition: A New Scaling Principle for Reliable Intelligence
- Oct 13, 2026Run fast, private agents on-device with LFM2.5-24B-A2B
- Oct 13, 2026Gemini 4 Argon for Robotics