Multi-Agent Reinforcement Learning framework for collaborative label aggregation on noisy classification datasets. Includes DQN agents, Gym-style environment, evaluation tools, and majority-voting baseline.
reinforcement-learning pytorch weak-supervision dqn multi-agent-systems data-annotation multi-agent-reinforcement-learning noisy-labels majority-voting label-aggregation cooperative-learning label-consensus crowdsourced-labels
-
Updated
Aug 3, 2025 - Python