Dyna-like reinforcement learning based on accumulative and average rewards.

Kao-Shing Hwang, Chia-Yue Lo

Browse the full SMC paper archive.