Skip to main content

Research Repository

Advanced Search

Guided Policy Search for Sequential Multitask Learning

Xiong, Fangzhou; Sun, Biao; Yang, Xu; Qiao, Hong; Huang, Kaizhu; Hussain, Amir; Liu, Zhiyong


Fangzhou Xiong

Biao Sun

Xu Yang

Hong Qiao

Kaizhu Huang

Zhiyong Liu


Policy search in reinforcement learning (RL) is a practical approach to interact directly with environments in parameter spaces, that often deal with dilemmas of local optima and real-time sample collection. A promising algorithm, known as guided policy search (GPS), is capable of handling the challenge of training samples using trajectory-centric methods. It can also provide asymptotic local convergence guarantees. However, in its current form, the GPS algorithm cannot operate in sequential multitask learning scenarios. This is due to its batch-style training requirement, where all training samples are collectively provided at the start of the learning process. The algorithm's adaptation is thus hindered for real-time applications, where training samples or tasks can arrive randomly. In this paper, the GPS approach is reformulated, by adapting a recently proposed, lifelong-learning method, and elastic weight consolidation. Specifically, Fisher information is incorporated to impart knowledge from previously learned tasks. The proposed algorithm, termed sequential multitask learning-GPS, is able to operate in sequential multitask learning settings and ensuring continuous policy learning, without catastrophic forgetting. Pendulum and robotic manipulation experiments demonstrate the new algorithms efficacy to learn control policies for handling sequentially arriving training samples, delivering comparable performance to the traditional, and batch-based GPS algorithm. In conclusion, the proposed algorithm is posited as a new benchmark for the real-time RL and robotics research community.

Journal Article Type Article
Acceptance Date Jan 11, 2018
Online Publication Date Feb 18, 2018
Publication Date 2019-01
Deposit Date Jul 19, 2019
Publicly Available Date Jul 19, 2019
Journal IEEE Transactions on Systems, Man, and Cybernetics: Systems
Print ISSN 2168-2216
Electronic ISSN 2168-2232
Publisher Institute of Electrical and Electronics Engineers
Peer Reviewed Peer Reviewed
Volume 49
Issue 1
Pages 216-226
Keywords Control and Systems Engineering; Human-Computer Interaction; Electrical and Electronic Engineering; Software; Computer Science Applications
Public URL
Related Public URLs
Contract Date Jul 19, 2019


You might also like

Downloadable Citations