Portal de Periódicos da CAPES

Proposal of a Deep Q-network with Profit Sharing

2018; Elsevier BV; Volume: 123; Linguagem: Inglês

10.1016/j.procs.2018.01.047

1877-0509

Kazuteru Miyazaki,

Artificial Intelligence in Games

Deep learning has attracted significant interest currently. The deep Q-network (DQN) combined with Q-learning have demonstrated excellent results for several Atari 2600 games. In this paper, we propose an exploitation-oriented learning (XoL) method that incorporates deep learning to reduce the number of trial-and-error searches. We focus on a profit sharing (PS) method that is an XoL method and combine a DQN and PS. The proposed method DQNwithPS is compared to a DQN in Pong of Atari 2600 games. We demonstrate that the proposed DQNwithPS method can learn stably with fewer trial-and-error searches than only using a DQN.