Zonglin Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Zonglin Liu has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Hanzhang Wang
School of Science, Harbin Institute of Technology (Shenzhen), Shenzhen, Guangdong, China
2 shared publications - Qiangqiang Shen
Department of Computer Science, City University of Hong Kong, Hong Kong, China
2 shared publications - Yongsheng Liang
College of Electronics and Information Engineering, Shenzhen University, Shenzhen, China
2 shared publications - Yongyong Chen
School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen, China
2 shared publications - Shiqi Wang
City University of Hong Kong, Hong Kong, China
1 shared publication - Zhiwei Zhong
Department of Computer Science, City University of Hong Kong, Kowloon, Hong Kong
1 shared publication
Latest Publications
- 21 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. PP (10) , 1-13
Linearized Tensor Low-rank Convolutional Network
To balance deep learning performance and interpretability, convolutional sparse coding (CSC) has successfully incorporated differentiable optimization layers in place of the standard convolutional layers of traditional deep neural networks (DNN). However, CSC models prefer to fall into local sparsity under the ℓ1-norm, whose optimization shows limitations in preserving the geometric structure and global sparsity of tensor data. In this paper, we propose a Linearized Tensor Low-Rank Convolutional Network (LTLRCNet) to refresh the differentiable optimization layers for DNN (e.g., ResNet and UNet), which instead optimizes the tensor low-rank minimization problem for global sparsity without ruining the intrinsic structure of data. Specifically, instead of simply replacing the ℓ1-norm with a tensor low-rank regularizer, we first propose a linearized tensor lowrank representation (LTLRR) model optimized by a linearized alternating direction method to decouple the dictionary and code coefficient without introducing an auxiliary variable, whose convergence is proved theoretically. Subsequently, we extend the LTLRR model into a deep unfolding network, such as the LTLRCNet, by engineering a convolutional dictionary that functions as a differentiable optimization layer, effectively substituting the conventional convolutional layer. To match the deep unfolding framework and avoid the gradient explosion, we apply a rank-one tactic bearing the Frank-Wolfe algorithm to solve the tensor low-rank minimization problem instead of conducting the cumbersome full-rank tensor singular value decomposition. Experimental results on image classification and segmentation show the potential of our LTLRCNet with its modeling-based layers to maintain significant performance and interpretability.
- 21 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. PP (10) , 1-13
Linearized Tensor Low-rank Convolutional Network
To balance deep learning performance and interpretability, convolutional sparse coding (CSC) has successfully incorporated differentiable optimization layers in place of the standard convolutional layers of traditional deep neural networks (DNN). However, CSC models prefer to fall into local sparsity under the ℓ1-norm, whose optimization shows limitations in preserving the geometric structure and global sparsity of tensor data. In this paper, we propose a Linearized Tensor Low-Rank Convolutional Network (LTLRCNet) to refresh the differentiable optimization layers for DNN (e.g., ResNet and UNet), which instead optimizes the tensor low-rank minimization problem for global sparsity without ruining the intrinsic structure of data. Specifically, instead of simply replacing the ℓ1-norm with a tensor low-rank regularizer, we first propose a linearized tensor lowrank representation (LTLRR) model optimized by a linearized alternating direction method to decouple the dictionary and code coefficient without introducing an auxiliary variable, whose convergence is proved theoretically. Subsequently, we extend the LTLRR model into a deep unfolding network, such as the LTLRCNet, by engineering a convolutional dictionary that functions as a differentiable optimization layer, effectively substituting the conventional convolutional layer. To match the deep unfolding framework and avoid the gradient explosion, we apply a rank-one tactic bearing the Frank-Wolfe algorithm to solve the tensor low-rank minimization problem instead of conducting the cumbersome full-rank tensor singular value decomposition. Experimental results on image classification and segmentation show the potential of our LTLRCNet with its modeling-based layers to maintain significant performance and interpretability.
- 7 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36 (4) , 4275-4286
Anchor-Induced Serial Tensor Representation for Multi-View Clustering
Multi-view clustering (MVC) has emerged as a powerful approach for integrating diverse sources of information from complex datasets. Nevertheless, existing methods struggle to accurately capture the global correlations and high-order structures in the data, and employ anchor-based techniques within a single dimension, limiting their representation. To address these issues, we propose an Anchor-induced Serial Tensor Representation (ASTR) framework, which effectively harnesses serial tensor representation to capture comprehensive multi-view information while reducing approximation errors and enhancing clustering performance. Specifically, ASTR begins with projection learning to explore low-dimensional latent spaces in multi-view data. Then, we introduce multi-anchor learning, where multiple anchor configurations are generated within the latent spaces, yielding a set of corresponding bipartite graphs. Besides, we organize these bipartite graphs into a sequence of global tensors, forming the serial tensor representation that encapsulates high-order inter- and intra-view relationships. Furthermore, we introduce the Laplace function to achieve a more accurate tensor rank approximation, complemented by a thorough theoretical analysis. Finally, a one-step clustering process, guided by adaptive weights, directly fuses the learned graphs to produce the final clustering indicator matrix. Experimental results demonstrate that ASTR possesses superior clustering accuracy and comparable efficiency.
- 7 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36 (4) , 4275-4286
Anchor-Induced Serial Tensor Representation for Multi-View Clustering
Multi-view clustering (MVC) has emerged as a powerful approach for integrating diverse sources of information from complex datasets. Nevertheless, existing methods struggle to accurately capture the global correlations and high-order structures in the data, and employ anchor-based techniques within a single dimension, limiting their representation. To address these issues, we propose an Anchor-induced Serial Tensor Representation (ASTR) framework, which effectively harnesses serial tensor representation to capture comprehensive multi-view information while reducing approximation errors and enhancing clustering performance. Specifically, ASTR begins with projection learning to explore low-dimensional latent spaces in multi-view data. Then, we introduce multi-anchor learning, where multiple anchor configurations are generated within the latent spaces, yielding a set of corresponding bipartite graphs. Besides, we organize these bipartite graphs into a sequence of global tensors, forming the serial tensor representation that encapsulates high-order inter- and intra-view relationships. Furthermore, we introduce the Laplace function to achieve a more accurate tensor rank approximation, complemented by a thorough theoretical analysis. Finally, a one-step clustering process, guided by adaptive weights, directly fuses the learned graphs to produce the final clustering indicator matrix. Experimental results demonstrate that ASTR possesses superior clustering accuracy and comparable efficiency.