Chunlin Chen

Dr. Chunlin Chen

Department of Control and Systems Engineering, Nanjing University, Nanjing, China

Research Focus & Metrics

Academic Career Length

21Years (2006-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 February 2018 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
ZRZhipeng Ren
Zhipeng Ren
DDDaoyi Dong
Daoyi Dong
HLHuaxiong Li
Huaxiong Li
CCChunlin Chen
Chunlin Chen
In this paper, a new training paradigm is proposed for deep reinforcement learning using self-paced prioritized curriculum learning with coverage penalty. The proposed deep curriculum reinforcement learning (DCRL) takes the most advantage of experience replay by adaptively selecting appropriate transitions from replay memory based on the complexity of each transition. The criteria of complexity in DCRL consist of self-paced priority as well as coverage penalty. The self-paced priority reflects the relationship between the temporal-difference error and the difficulty of the current curriculum for sample efficiency. The coverage penalty is taken into account for sample diversity. With comparison to deep Q network (DQN) and prioritized experience replay (PER) methods, the DCRL algorithm is evaluated on Atari 2600 games, and the experimental results show that DCRL outperforms DQN and PER on most of these games. More results further show that the proposed curriculum training paradigm of DCRL is also applicable and effective for other memory-based deep reinforcement learning approaches, such as double DQN and dueling network. All the experimental results demonstrate that DCRL can achieve improved training efficiency and robustness for deep reinforcement learning.

18 February 2021 in IEEE Transactions on Cybernetics
Verified
Journal Article
Closed access
QWQing Wei
Qing Wei
HMHailan Ma
Hailan Ma
CCChunlin Chen
Chunlin Chen
DDDaoyi Dong
Daoyi Dong
In this article, a novel training paradigm inspired by quantum computation is proposed for deep reinforcement learning (DRL) with experience replay. In contrast to the traditional experience replay mechanism in DRL, the proposed DRL with quantum-inspired experience replay (DRL-QER) adaptively chooses experiences from the replay buffer according to the complexity and the replayed times of each experience (also called transition), to achieve a balance between exploration and exploitation. In DRL-QER, transitions are first formulated in quantum representations and then the preparation operation and depreciation operation are performed on the transitions. In this process, the preparation operation reflects the relationship between the temporal-difference errors (TD-errors) and the importance of the experiences, while the depreciation operation is taken into account to ensure the diversity of the transitions. The experimental results on Atari 2600 games show that DRL-QER outperforms state-of-the-art algorithms, such as DRL-PER and DCRL on most of these games with improved training efficiency and is also applicable to such memory-based DRL approaches as double network and dueling network.

10 July 2019 in IEEE Transactions on Cybernetics
Verified
Journal Article
Open access
DDDaoyi Dong
Daoyi Dong
XXXi Xing
Xi Xing
HMHailan Ma
Hailan Ma
CCChunlin Chen
Chunlin Chen
ZLZhixin Liu
Zhixin Liu
HRHershel Rabitz
Hershel Rabitz
Robust control design for quantum systems has been recognized as a key task in quantum information technology, molecular chemistry, and atomic physics. In this paper, an improved differential evolution algorithm, referred to as multiple-samples and mixed-strategy DE (msMS_DE), is proposed to search robust fields for various quantum control problems. In msMS_DE, multiple samples are used for fitness evaluation and a mixed strategy is employed for the mutation operation. In particular, the msMS_DE algorithm is applied to the control problems of: 1) open inhomogeneous quantum ensembles and 2) the consensus goal of a quantum network with uncertainties. Numerical results are presented to demonstrate the excellent performance of the improved machine learning algorithm for these two classes of quantum robust control problems. Furthermore, msMS_DE is experimentally implemented on femtosecond (fs) laser control applications to optimize two-photon absorption and control fragmentation of the molecule CH₂BrI. The experimental results demonstrate the excellent performance of msMS_DE in searching for effective fs laser pulses for various tasks.

19 January 2015 in Scientific Reports
Verified
Journal Article
Open access
PDF file
DDDaoyi Dong
Daoyi Dong
CCChunlin Chen
Chunlin Chen
BQBo Qi
Bo Qi
IPIan R. Petersen
Ian R. Petersen
FNFranco Nori
Franco Nori
Superconducting quantum systems are promising candidates for quantum information processing due to their scalability and design flexibility. However, the existence of defects, fluctuations, and inaccuracies is unavoidable for practical superconducting quantum circuits. In this paper, a sampling-based learning control (SLC) method is used to guide the design of control fields for manipulating superconducting quantum systems. Numerical results for one-qubit systems and coupled two-qubit systems show that the “smart” fields learned using the SLC method can achieve robust manipulation of superconducting qubits, even in the presence of large fluctuations and inaccuracies.

1 January 2014 in Mathematical Problems in Engineering
Verified
Journal Article
Open access
PDF file
DXDongdong Xu
Dongdong Xu
XZXingnan Zhang
Xingnan Zhang
ZZZhangqing Zhu
Zhangqing Zhu
CCChunlin Chen
Chunlin Chen
PYPei Yang
Pei Yang
Swarm robotics is a specific research field of multirobotics where a large number of mobile robots are controlled in a coordinated way. Formation control is one of the most challenging goals for the coordination control of swarm robots. In this paper, a behavior‐based control design approach is proposed for two kinds of important formation control problems: efficient initial formation and formation control while avoiding obstacles. In this approach, a classification‐based searching method for generating large‐scale robot formation is presented to reduce the computational complexity and speed up the initial formation process for any desired formation. The behavior‐based method is applied for the formation control of swarm robot systems while navigating in an unknown environment with obstacles. Several groups of experimental results demonstrate the success of the proposed approach. These methods have potential applications for various swarm robot systems in both the simulation and the practical environments.

18 March 2019 in IEEE Transactions on Cybernetics
Verified
Journal Article
Closed access
ZWZhi Wang
Zhi Wang
HLHan-Xiong Li
Han-Xiong Li
CCChunlin Chen
Chunlin Chen
A reinforcement learning-based method is proposed for optimal sensor placement in the spatial domain for modeling distributed parameter systems (DPSs). First, a low-dimensional subspace, derived by Karhunen-Loève decomposition, is identified to capture the dominant dynamic features of the DPS. Second, a spatial objective function is proposed for the sensor placement. This function is defined in the obtained low-dimensional subspace by exploiting the time-space separation property of distributed processes, and in turn aims at minimizing the modeling error over the entire time and space domain. Third, the sensor placement configuration is mathematically formulated as a Markov decision process (MDP) with specified elements. Finally, the sensor locations are optimized through learning the optimal policies of the MDP according to the spatial objective function. The experimental results of a simulated catalytic rod and a real snap curing oven system are provided to demonstrate the feasibility and efficiency of the proposed method in solving the combinatorial optimization problems, such as optimal sensor placement.