The perfect score for arcade classic Ms. Pac-Man has been achieved. Maluuba — a deep learning team acquired by Microsoft — has created an AI system that’s learned how to reach the game’s maximum point value of 999,900 on Atari 2600, using a unique combination of reinforcement learning with a divide-and-conquer method.
Although AI has conquered a number of retro games, Ms. Pac-Man has remained elusive, due to the game’s lack of predictability. Maluuba tasked out responsibilities to over 150 agents. The team then taught the AI using what they call Hybrid Reward Architecture — a combination of reinforcement learning with a divide-and-conquer method. Individual agents were assigned piecemeal tasks — like finding a specific pellet before working in tandem with other agents to achieve greater goals. Maluuba then designated a top AI agent that took suggestions from the other agents in order to inform decisions on where to move Ms. Pac-Man.
The best results came when individual agents “acted very egotistically” and the top agent focused on what was best for the overall team, taking into account not only how many agents wanted to go in a particular direction, but the importance of that direction. (Example: fewer agents wanting to avoid a ghost took priority over a higher amount of agents wanting to pursue a pellet.) “There’s this nice interplay,” says Harm Van Seijen, a researcher with Maluuba, “between how they have to, on the one hand, cooperate based on the preferences of all the agents, but at the same time each agent cares only about one particular problem. It benefits the whole.”
------------
No comments:
Post a Comment