A role-based multi-agent reinforcement learning system with mixture-of-expert policies improves simulated active object tracking episode length over a single-agent baseline, though only modestly and without ablation or code.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
CSAOT: Cooperative Multi-Agent System for Active Object Tracking
A role-based multi-agent reinforcement learning system with mixture-of-expert policies improves simulated active object tracking episode length over a single-agent baseline, though only modestly and without ablation or code.