Mei Liu

Mei Liu

Auto-generated profile, based on publicly available information

Mei Liu

Multi-Scale Medical Robotics Center, The Chinese University of Hong Kong, Hong Kong, 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 September 2021 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
MLMei Liu
Mei Liu
LCLiangming Chen
Liangming Chen
XDXiaohao Du
Xiaohao Du
Long JinLong Jin
Long Jin
MSMingsheng Shang
Mingsheng Shang
Deep neural networks often suffer from poor performance or even training failure due to the ill-conditioned problem, the vanishing/exploding gradient problem, and the saddle point problem. In this article, a novel method by acting the gradient activation function (GAF) on the gradient is proposed to handle these challenges. Intuitively, the GAF enlarges the tiny gradients and restricts the large gradient. Theoretically, this article gives conditions that the GAF needs to meet and, on this basis, proves that the GAF alleviates the problems mentioned above. In addition, this article proves that the convergence rate of SGD with the GAF is faster than that without the GAF under some assumptions. Furthermore, experiments on CIFAR, ImageNet, and PASCAL visual object classes confirm the GAF's effectiveness. The experimental results also demonstrate that the proposed method is able to be adopted in various deep neural networks to improve their performance. The source code is publicly available at https://github.com/LongJin-lab/Activated-Gradients-for-Deep-Neural-Networks.

7 May 2024 in IEEE Transactions on Mobile Computing
Verified
Journal Article
Closed access
MLMei Liu
Mei Liu
YLYutong Li
Yutong Li
YCYingqi Chen
Yingqi Chen
YQYimeng Qi
Yimeng Qi
Long JinLong Jin
Long Jin
Enlightened by competitive and collaborative coordination behaviors widely observed in natural swarm systems, this work emphasizes these coordinating modes in multirobot systems and optimizes system stability along with resource utilization. Then, schemes are constructed to describe and model these two modes, where a $k$ -winner-take-all concept is introduced as the driving principle of multirobot competition. In addition, a distributed coordination approach is established to effectively handle the above schemes aided with optimality theory, which is developed by a fusion of a recurrent neural dynamics solver and a distributed solver. The former is a single-layer neural dynamics model with a simple structure, and the latter transforms the involved global information to a distributed type via consensus. Both of them are carried out in the discrete-time domain to fit the actual application. Finally, the convergence and stability of the proposed coordination approach are proved via theoretical analysis and further demonstrated through simulations and experiments.

20 October 2020 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
ZXZhengtai Xie
Zhengtai Xie
Long JinLong Jin
Long Jin
XWXin DI WU
Xin DI WU
ZSZhongbo Sun
Zhongbo Sun
MLMei Liu
Mei Liu
For the existing repetitive motion generation (RMG) schemes for kinematic control of redundant manipulators, the position error always exists and fluctuates. This article gives an answer to this phenomenon and presents the theoretical analyses to reveal that the existing RMG schemes exist a theoretical position error related to the joint angle error. To remedy this weakness of existing solutions, an orthogonal projection RMG (OPRMG) scheme is proposed in this article by introducing an orthogonal projection method with the position error eliminated theoretically, which decouples the joint space error and Cartesian space error with joint constraints considered. The corresponding new recurrent neural networks (NRNNs) are structured by exploiting the gradient descent method with the assistance of velocity compensation with theoretical analyses provided to embody the stability and feasibility. In addition, simulation results on a fixed-based redundant manipulator, a mobile manipulator, and a multirobot system synthesized by the existing RMG schemes and the proposed one are presented to verify the superiority and precise performance of the OPRMG scheme for kinematic control of redundant manipulators. Moreover, via adjusting the coefficient, simulations on the position error and joint drift of the redundant manipulator are conducted for comparison to prove the high performance of the OPRMG scheme. To bring out the crucial point, different controllers for the redundancy resolution of redundant manipulators are compared to highlight the superiority and advantage of the proposed NRNN. This work greatly improves the existing RMG solutions in theoretically eliminating the position error and joint drift, which is of significant contributions to increasing the accuracy and efficiency of high-precision instruments in manufacturing production.

16 June 2021 in IEEE Transactions on Cybernetics
Verified
Journal Article
Closed access
YQYimeng Qi
Yimeng Qi
Long JinLong Jin
Long Jin
XWXin DI WU
Xin DI WU
YSYang Shi
Yang Shi
MLMei Liu
Mei Liu
In this article, a robust $k$ -winner-take-all ( $k$ -WTA) neural network employing the saturation-allowed activation functions is designed and investigated to perform a $k$ -WTA operation, and is shown to possess enhanced robustness to disturbance compared to existing $k$ -WTA neural networks. Global convergence and robustness of the proposed $k$ -WTA neural network are demonstrated through analysis and simulations. An application studied in detail is competitive multiagent coordination and dynamic task allocation, in which $k$ active agents [among $m~(m > k)$ ] are allocated to execute a tracking task with the static $m-k$ ones. This is implemented by adopting a distributed $k$ -WTA network with limited communication, aided with a consensus filter. Simulation results demonstrating the system’s efficacy and feasibility are presented.

26 April 2023 in Neural Networks
Verified
Journal Article
Closed access
CQChuan Qin
Chuan Qin
LCLiangming Chen
Liangming Chen
ZCZangtai Cai
Zangtai Cai
MLMei Liu
Mei Liu
Long JinLong Jin
Long Jin

10 July 2020 in IEEE Transactions on Industrial Electronics
Verified
Journal Article
Closed access
Long JinLong Jin
Long Jin
JZJiazheng Zhang
Jiazheng Zhang
XWXin DI WU
Xin DI WU
MLMei Liu
Mei Liu
Shuai  LiShuai Li
Shuai Li
ZYZihao Yang
Zihao Yang
For avoiding the singularity arising in the cooperative control of multiple redundant robot manipulators, an efficient way is to maximize the manipulability. In this paper, by making progress along this direction, a distributed manipulability optimization scheme is proposed to maximize the manipulability of redundant robot manipulators in a distributed network with limited communication. With manipulability optimization incorporated in the proposed scheme, all the involved manipulators can be regulated to track their optimal configurations dynamically, in addition to the collaboration among them to complete the specified tasks. To do this, the distributed scheme is transformed into a dynamic quadratic programming (QP) problem by considering the time dependence of the parameters. Then, a generalized recurrent neural network (GRNN) is constructed and proposed to deal with the QP problem online with perturbations considered. Theoretical analysis is conducted, which confirms that the proposed GRNN is able to globally converge to the optimal solution to the dynamic QP problem in the presence of noises and perturbations. Finally, simulation results based on a distributed network of redundant robots are conducted and presented to verify the superior performance of the proposed distributed manipulability optimization scheme.