Fangzhou Xiong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Fangzhou Xiong has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Amir Hussain
School of Computing, Edinburgh Napier University, Scotland, UK
2 shared publications - Kaizhu Huang
Duke Kunshan University, Kunshan, China
2 shared publications - Zhiyong Liu
Institute of Automation, Chinese Academy of Sciences
2 shared publications - Biao Sun
School of Automation and Electrical Engineering, University of Science and Technology Beijing, Beijing 100083, China
1 shared publication Hong Qiao
State Key Lab of Management and Control for Complex Systems, Institute of Automation, Chinese Academy of Science, 100190, Beijing, China; School of Artificial Intelligence, University of Chinese Academy of Sciences (UCAS), 100049, Beijing, China; CAS Centre for Excellence in Brain Science and Intelligence Technology, Chinese Academy of Sciences, 200031, Shanghai, China
1 shared publication- Hong Qiao
Institute of Automation, Chinese Academy of Sciences
1 shared publication
Latest Publications
- 1 September 2020
- Published by Elsevier in Neural Networks
- Vol. 129, 163-173
Encoding primitives generation policy learning for robotic arm to overcome catastrophic forgetting in sequential multi-tasks learning
- 1 September 2020
- Published by Elsevier in Neural Networks
- Vol. 129, 163-173
Encoding primitives generation policy learning for robotic arm to overcome catastrophic forgetting in sequential multi-tasks learning
- 19 February 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Systems, Man, and Cybernetics: Systems
- Vol. 49 (1) , 216-226
Guided Policy Search for Sequential Multitask Learning
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.
- 19 February 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Systems, Man, and Cybernetics: Systems
- Vol. 49 (1) , 216-226
Guided Policy Search for Sequential Multitask Learning
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.