Google Research Football (GRF)

The 'Google Research Football (GRF)' is a benchmark that contains a simulated football environment used to evaluate multi-agent reinforcement learning algorithms in terms of their ability to coordinate and collaborate effectively.

Papers using Google Research Football (GRF) (11)

Adaptive Context Length Optimization with Low-Frequency Truncation for Multi-Agent Reinforcement Learning2025 · 3 citesCCKS: Consensus-based Communication and Knowledge Sharing2026SPECTra: Scalable Multi-Agent Reinforcement Learning with Permutation-Free Networks2025 · 2 citesDual-Gated Epistemic Time-Dilation: Autonomous Compute Modulation in Asynchronous MARL2026Autonomous Partner Selection for Cooperative Multi-Agent Reinforcement Learning2026Bridging MARL to SARL: An Order-Independent Multi-Agent Transformer via Latent Consensus2026Multi-Agent Deep Reinforcement Learning Under Constrained Communications2026LAGMA: Latent Goal-guided Multi-agent Reinforcement Learning2024 · 1 citesLearning to Collaborate by Grouping: a Consensus-oriented Strategy for Multi-agent Reinforcement Learning2023 · 7 citesPTDE: Personalized Training with Distilled Execution for Multi-Agent Reinforcement Learning2022 · 4 citesMulti-Task Multi-Agent Shared Layers are Universal Cognition of Multi-Agent Coordination2023