Composable Modular Reinforcement Learning

Simpkins, C; Isbell, C

Simpkins, C (reprint author), Georgia Inst Technol, Coll Comp, 801 Atlantic Dr, Atlanta, GA 30332 USA.

THIRTY-THIRD AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE / THIRTY-FIRST INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE / NINTH, 2019; (): 4975

Abstract

Modular reinforcement learning (MRL) decomposes a monolithic multiple-goal problem into modules that solve a portion of the original problem. The modu......

Full Text Link