Japan saw the NeurIPS PokeAgent Challenge and said, "pokemon trading card game next"
Many of the training algorithms, models, and infra takeaways from the challenge should be useful for bootstrapping a TCG agent
Megathread for previous work that can help participants 👇