References & Citations
Computer Science > Artificial Intelligence
Title: Driving Reinforcement Learning with Models
(Submitted on 11 Nov 2019 (v1), last revised 26 Feb 2020 (this version, v2))
Abstract: In this paper we propose a new approach to complement reinforcement learning (RL) with model-based control (in particular, Model Predictive Control - MPC). We introduce an algorithm, the MPC augmented RL (MPRL) that combines RL and MPC in a novel way so that they can augment each other's strengths. We demonstrate the effectiveness of the MPRL by letting it play against the Atari game Pong. For this task, the results highlight how MPRL is able to outperform both RL and MPC when these are used individually.
Submission history
From: Pietro Ferraro [view email][v1] Mon, 11 Nov 2019 17:14:56 GMT (767kb,D)
[v2] Wed, 26 Feb 2020 15:25:52 GMT (1456kb,D)
Link back to: arXiv, form interface, contact.