DQN In this project, an artificial agent is trained to solve the Maze task using DQN algorithm game.mov Below figure shows the optimal policy found by the agent: Below figure shows the learning curve:
