GRF

The 'GRF' dataset/benchmark is used to evaluate multi-agent reinforcement learning (MARL) methods by providing scenarios that assess the effectiveness of strategies promoting diversity among agents through competitive and cooperative interactions.

Papers using GRF (3)

An Extended Benchmarking of Multi-Agent Reinforcement Learning Algorithms in Complex Fully Cooperative Tasks2025 · 1 citesConstructive Conflict-Driven Multi-Agent Reinforcement Learning for Strategic Diversity2025Intrinsic Action Tendency Consistency for Cooperative Multi-Agent Reinforcement Learning2024