GATAis trained using acombination of reinforcement and self-supervised learning. Our workdemonstrates thatthelearned graph-based representations helpagents converge to better policies than their text-only counterparts and facilitate effective generalization across game configurations.

Updating onlythese task-specific modules thenallowsthe model to be adapted to low-data tasks for as many steps as necessary without risking overfitting. Unfortunately, existing meta-learning methods either do not scale to long adaptation or else rely on handcrafted task-specific architectures. Here, we propose ameta-learning approach that obviates the need for this often sub-optimal hand-selection.