This is a Taxi-Pickup game (which is inspired and a remake of the OpenAI's Gym/Gymnasium TaxiV3 game for Reinforcement Learning Tutorials). It uses a simple Q-learning table backend to simulate a possible Reinforcement Learning agent as AI-powered NPC for on-chain games (if combo-ed with ZKML like Giza and EZKL).
The backend has to be powered by our Infra Hack's main Auto-Topper contract here and Paymaster simulation here.
