Shao-Qun Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Shao-Qun Zhang has a total of 24 co-authors. Here are the top co-authors by number of shared publications:
- Zhi-Hua Zhou
Nanjing University
10 shared publications - Yuan Jiang
College of Materials, MOE Key Laboratory of High-Performance Ceramic Fibers, Fujian Key Laboratory of Advanced Materials, Higher Educational Key Laboratory for Biomedical Engineering of Fujian Province, Xiamen University, Xiamen 361005, P. R. China
5 shared publications - Gao Zhang
Institute of Mathematics Nanjing Normal University China
3 shared publications - Feng-Lei Fan
Department of Data Science, City University of Hong Kong, Hong Kong, China
2 shared publications Jin-Hui Wu
National Key Laboratory for Novel Software Technology Nanjing University China
2 shared publications- Bin Gu
Department of Machine Learning Mohamed Zayed University of Artificial Intelligence Abu Dhabi UAE
1 shared publication
Latest Publications
- 8 October 2026
- Published by Springer Nature in Machine Learning
- Vol. 115 (10) , 242
Grid-based Decision Tree Learning
- 8 October 2026
- Published by Springer Nature in Machine Learning
- Vol. 115 (10) , 242
Grid-based Decision Tree Learning
- Open Access
TernaryCLIP: Efficiently Compressing Vision-Language Models with Ternary Weights and Distilled Knowledge
Recent years have witnessed an increasing interest in image-text contrastive modeling, exemplified by models such as Contrastive Language-Image Pretraining (CLIP). In this paper, we propose the TernaryCLIP, a lightweight computational framework that converts connection weights of both vision and text encoders of CLIP into the ternary format, instead of full-precision or floating ones. TernaryCLIP incorporates quantization-aware training and distillation modules, preventing precision degradation and enabling low-cost and high-efficiency computations. Comprehensive experiments demonstrate that TernaryCLIP can achieve up to 99\% ternarized weights with 1.58-bit representation, 16.98 $\times$ compression ratio, 2.3 $\times$ inference acceleration, 16 $\times$ storage reduction, 10 $\times$ memory optimization, and 60\% sparsity while maintaining promising performance on zero-shot image classification and image-text retrieval tasks across 41 commonly used datasets. Our work highlights the feasibility of extreme quantization for large multimodal models, supporting effective and efficient deployment on resource-constrained devices. The model and code can be accessed from Hugging Face and GitHub.Open AccessTernaryCLIP: Efficiently Compressing Vision-Language Models with Ternary Weights and Distilled Knowledge
Recent years have witnessed an increasing interest in image-text contrastive modeling, exemplified by models such as Contrastive Language-Image Pretraining (CLIP). In this paper, we propose the TernaryCLIP, a lightweight computational framework that converts connection weights of both vision and text encoders of CLIP into the ternary format, instead of full-precision or floating ones. TernaryCLIP incorporates quantization-aware training and distillation modules, preventing precision degradation and enabling low-cost and high-efficiency computations. Comprehensive experiments demonstrate that TernaryCLIP can achieve up to 99\% ternarized weights with 1.58-bit representation, 16.98 $\times$ compression ratio, 2.3 $\times$ inference acceleration, 16 $\times$ storage reduction, 10 $\times$ memory optimization, and 60\% sparsity while maintaining promising performance on zero-shot image classification and image-text retrieval tasks across 41 commonly used datasets. Our work highlights the feasibility of extreme quantization for large multimodal models, supporting effective and efficient deployment on resource-constrained devices. The model and code can be accessed from Hugging Face and GitHub. - Open Access
- 24 March 2024
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 38 (10) , 11078-11086
MEPSI: An MDL-Based Ensemble Pruning Approach with Structural Information
Ensemble pruning that combines a subset of individual learners generated in parallel to make predictions is an important topic in ensemble learning. Past decades have developed a lot of pruning algorithms that focus on the external behavior of learners on samples, which may lead to over-fitting. In this paper, we conjecture that the generalization performance of an ensemble is not only related to its external behavior on samples but also dependent on the internal structure of individual learners. We propose the general MEPSI approach based on Kolmogorov complexity and the Minimum Description Length (MDL) principle, which formulates the ensemble pruning task as the two-objective optimization problem that comprises the empirical error and structural information among individual learners. We also provide a concrete implementation of MEPSI on decision trees. The theoretical results provide generalization bounds for both the general MEPSI approach and tree-based implementation. The comparative experiments conducted on multiple real-world data sets demonstrate the effectiveness of our proposed method.
- 24 March 2024
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 38 (10) , 11078-11086
Open AccessMEPSI: An MDL-Based Ensemble Pruning Approach with Structural Information
Ensemble pruning that combines a subset of individual learners generated in parallel to make predictions is an important topic in ensemble learning. Past decades have developed a lot of pruning algorithms that focus on the external behavior of learners on samples, which may lead to over-fitting. In this paper, we conjecture that the generalization performance of an ensemble is not only related to its external behavior on samples but also dependent on the internal structure of individual learners. We propose the general MEPSI approach based on Kolmogorov complexity and the Minimum Description Length (MDL) principle, which formulates the ensemble pruning task as the two-objective optimization problem that comprises the empirical error and structural information among individual learners. We also provide a concrete implementation of MEPSI on decision trees. The theoretical results provide generalization bounds for both the general MEPSI approach and tree-based implementation. The comparative experiments conducted on multiple real-world data sets demonstrate the effectiveness of our proposed method.
- Open Access
On the Intrinsic Structures of Spiking Neural Networks
Recent years have emerged a surge of interest in SNNs owing to their remarkable potential to handle time-dependent and event-driven data. The performance of SNNs hinges not only on selecting an apposite architecture and fine-tuning connection weights, similar to conventional ANNs, but also on the meticulous configuration of intrinsic structures within spiking computations. However, there has been a dearth of comprehensive studies examining the impact of intrinsic structures. Consequently, developers often find it challenging to apply a standardized configuration of SNNs across diverse datasets or tasks. This work delves deep into the intrinsic structures of SNNs. Initially, we unveil two pivotal components of intrinsic structures: the integration operation and firing-reset mechanism, by elucidating their influence on the expressivity of SNNs. Furthermore, we draw two key conclusions: the membrane time hyper-parameter is intimately linked to the eigenvalues of the integration operation, dictating the functional topology of spiking dynamics, and various hyper-parameters of the firing-reset mechanism govern the overall firing capacity of an SNN, mitigating the injection ratio or sampling density of input data. These findings elucidate why the efficacy of SNNs hinges heavily on the configuration of intrinsic structures and lead to a recommendation that enhancing the adaptability of these structures contributes to improving the overall performance and applicability of SNNs. Inspired by this recognition, we propose two feasible approaches to enhance SNN learning. These involve leveraging self-connection architectures and employing stochastic spiking neurons to augment the adaptability of the integration operation and firing-reset mechanism, respectively. We verify the effectiveness of the proposed methods from perspectives of theory and practice.
Open AccessOn the Intrinsic Structures of Spiking Neural Networks
Recent years have emerged a surge of interest in SNNs owing to their remarkable potential to handle time-dependent and event-driven data. The performance of SNNs hinges not only on selecting an apposite architecture and fine-tuning connection weights, similar to conventional ANNs, but also on the meticulous configuration of intrinsic structures within spiking computations. However, there has been a dearth of comprehensive studies examining the impact of intrinsic structures. Consequently, developers often find it challenging to apply a standardized configuration of SNNs across diverse datasets or tasks. This work delves deep into the intrinsic structures of SNNs. Initially, we unveil two pivotal components of intrinsic structures: the integration operation and firing-reset mechanism, by elucidating their influence on the expressivity of SNNs. Furthermore, we draw two key conclusions: the membrane time hyper-parameter is intimately linked to the eigenvalues of the integration operation, dictating the functional topology of spiking dynamics, and various hyper-parameters of the firing-reset mechanism govern the overall firing capacity of an SNN, mitigating the injection ratio or sampling density of input data. These findings elucidate why the efficacy of SNNs hinges heavily on the configuration of intrinsic structures and lead to a recommendation that enhancing the adaptability of these structures contributes to improving the overall performance and applicability of SNNs. Inspired by this recognition, we propose two feasible approaches to enhance SNN learning. These involve leveraging self-connection architectures and employing stochastic spiking neurons to augment the adaptability of the integration operation and firing-reset mechanism, respectively. We verify the effectiveness of the proposed methods from perspectives of theory and practice.
- Open AccessLax Extensions of Conical -Semifilter MonadsFor a quantale the unit interval endowed with a continuous triangular norm, we introduce the canonical, op-canonical and Kleisli extensions of the conical -semifilter monad to It is proved that the op-canonical extension coincides with the Kleisli extension.Open AccessLax Extensions of Conical -Semifilter MonadsFor a quantale the unit interval endowed with a continuous triangular norm, we introduce the canonical, op-canonical and Kleisli extensions of the conical -semifilter monad to It is proved that the op-canonical extension coincides with the Kleisli extension.
- 26 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 35 (3) , 3674-3688
Theoretical Exploration of Flexible Transmitter Model
Neural network models generally involve two important components, i.e., network architecture and neuron model. Although there are abundant studies about network architectures, only a few neuron models have been developed, such as the MP neuron model developed in 1943 and the spiking neuron model developed in the 1950s. Recently, a new bio-plausible neuron model, flexible transmitter (FT) model (Zhang and Zhou, 2021), has been proposed. It exhibits promising behaviors, particularly on temporal–spatial signals, even when simply embedded into the common feedforward network architecture. This article attempts to understand the properties of the FT network (FTNet) theoretically. Under mild assumptions, we show that: 1) FTNet is a universal approximator; 2) the approximation complexity of FTNet can be exponentially smaller than those of commonly used real-valued neural networks with feedforward/recurrent architectures and is of the same order in the worst case; and 3) any local minimum of FTNet is the global minimum, implying that it is possible to identify global minima by local search algorithms.
- 26 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 35 (3) , 3674-3688
Theoretical Exploration of Flexible Transmitter Model
Neural network models generally involve two important components, i.e., network architecture and neuron model. Although there are abundant studies about network architectures, only a few neuron models have been developed, such as the MP neuron model developed in 1943 and the spiking neuron model developed in the 1950s. Recently, a new bio-plausible neuron model, flexible transmitter (FT) model (Zhang and Zhou, 2021), has been proposed. It exhibits promising behaviors, particularly on temporal–spatial signals, even when simply embedded into the common feedforward network architecture. This article attempts to understand the properties of the FT network (FTNet) theoretically. Under mild assumptions, we show that: 1) FTNet is a universal approximator; 2) the approximation complexity of FTNet can be exponentially smaller than those of commonly used real-valued neural networks with feedforward/recurrent architectures and is of the same order in the worst case; and 3) any local minimum of FTNet is the global minimum, implying that it is possible to identify global minima by local search algorithms.
- Open Access
On Discrete Presheaf Monads
For a quantale , which is a unit interval endowed with a continuous triangular norm and the Barr extension of the ultrafilter monad to a characterization of the discrete presheaf monad associated to is given. It is also proved that, when & is the Łucasiewicz triangular norm, the discrete presheaf monad is isomorphic to the saturated prefilter monad, and when & is the product triangular norm, the prime functional ideal monad is isomorphic to a submonad of the discrete presheaf monad.Open AccessOn Discrete Presheaf Monads
For a quantale , which is a unit interval endowed with a continuous triangular norm and the Barr extension of the ultrafilter monad to a characterization of the discrete presheaf monad associated to is given. It is also proved that, when & is the Łucasiewicz triangular norm, the discrete presheaf monad is isomorphic to the saturated prefilter monad, and when & is the product triangular norm, the prime functional ideal monad is isomorphic to a submonad of the discrete presheaf monad. - Open Access
Theoretical Exploration of Flexible Transmitter Model
Neural network models generally involve two important components, i.e., network architecture and neuron model. Although there are abundant studies about network architectures, only a few neuron models have been developed, such as the MP neuron model developed in 1943 and the spiking neuron model developed in the 1950s. Recently, a new bio-plausible neuron model, Flexible Transmitter (FT) model, has been proposed. It exhibits promising behaviors, particularly on temporal-spatial signals, even when simply embedded into the common feedforward network architecture. This paper attempts to understand the properties of the FT network (FTNet) theoretically. Under mild assumptions, we show that: i) FTNet is a universal approximator; ii) the approximation complexity of FTNet can be exponentially smaller than those of commonly-used real-valued neural networks with feedforward/recurrent architectures and is of the same order in the worst case; iii) any local minimum of FTNet is the global minimum, implying that it is possible to identify global minima by local search algorithms.
Open AccessTheoretical Exploration of Flexible Transmitter Model
Neural network models generally involve two important components, i.e., network architecture and neuron model. Although there are abundant studies about network architectures, only a few neuron models have been developed, such as the MP neuron model developed in 1943 and the spiking neuron model developed in the 1950s. Recently, a new bio-plausible neuron model, Flexible Transmitter (FT) model, has been proposed. It exhibits promising behaviors, particularly on temporal-spatial signals, even when simply embedded into the common feedforward network architecture. This paper attempts to understand the properties of the FT network (FTNet) theoretically. Under mild assumptions, we show that: i) FTNet is a universal approximator; ii) the approximation complexity of FTNet can be exponentially smaller than those of commonly-used real-valued neural networks with feedforward/recurrent architectures and is of the same order in the worst case; iii) any local minimum of FTNet is the global minimum, implying that it is possible to identify global minima by local search algorithms.
- Open Access
Neural Network Gaussian Processes by Increasing Depth
Recent years have witnessed an increasing interest in the correspondence between infinitely wide networks and Gaussian processes. Despite the effectiveness and elegance of the current neural network Gaussian process theory, to the best of our knowledge, all the neural network Gaussian processes are essentially induced by increasing width. However, in the era of deep learning, what concerns us more regarding a neural network is its depth as well as how depth impacts the behaviors of a network. Inspired by a width-depth symmetry consideration, we use a shortcut network to show that increasing the depth of a neural network can also give rise to a Gaussian process, which is a valuable addition to the existing theory and contributes to revealing the true picture of deep learning. Beyond the proposed Gaussian process by depth, we theoretically characterize its uniform tightness property and the smallest eigenvalue of the Gaussian process kernel. These characterizations can not only enhance our understanding of the proposed depth-induced Gaussian process but also pave the way for future applications. Lastly, we examine the performance of the proposed Gaussian process by regression experiments on two benchmark data sets.
Open AccessNeural Network Gaussian Processes by Increasing Depth
Recent years have witnessed an increasing interest in the correspondence between infinitely wide networks and Gaussian processes. Despite the effectiveness and elegance of the current neural network Gaussian process theory, to the best of our knowledge, all the neural network Gaussian processes are essentially induced by increasing width. However, in the era of deep learning, what concerns us more regarding a neural network is its depth as well as how depth impacts the behaviors of a network. Inspired by a width-depth symmetry consideration, we use a shortcut network to show that increasing the depth of a neural network can also give rise to a Gaussian process, which is a valuable addition to the existing theory and contributes to revealing the true picture of deep learning. Beyond the proposed Gaussian process by depth, we theoretically characterize its uniform tightness property and the smallest eigenvalue of the Gaussian process kernel. These characterizations can not only enhance our understanding of the proposed depth-induced Gaussian process but also pave the way for future applications. Lastly, we examine the performance of the proposed Gaussian process by regression experiments on two benchmark data sets.
- 4 July 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 35 (2) , 2881-2886
Neural Network Gaussian Processes by Increasing Depth
Recent years have witnessed an increasing interest in the correspondence between infinitely wide networks and Gaussian processes. Despite the effectiveness and elegance of the current neural network Gaussian process theory, to the best of our knowledge, all the neural network Gaussian processes (NNGPs) are essentially induced by increasing width. However, in the era of deep learning, what concerns us more regarding a neural network is its depth as well as how depth impacts the behaviors of a network. Inspired by a width-depth symmetry consideration, we use a shortcut network to show that increasing the depth of a neural network can also give rise to a Gaussian process, which is a valuable addition to the existing theory and contributes to revealing the true picture of deep learning. Beyond the proposed Gaussian process by depth, we theoretically characterize its uniform tightness property and the smallest eigenvalue of the Gaussian process kernel. These characterizations can not only enhance our understanding of the proposed depth-induced Gaussian process but also pave the way for future applications. Lastly, we examine the performance of the proposed Gaussian process by regression experiments on two benchmark datasets.
- 4 July 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 35 (2) , 2881-2886
Neural Network Gaussian Processes by Increasing Depth
Recent years have witnessed an increasing interest in the correspondence between infinitely wide networks and Gaussian processes. Despite the effectiveness and elegance of the current neural network Gaussian process theory, to the best of our knowledge, all the neural network Gaussian processes (NNGPs) are essentially induced by increasing width. However, in the era of deep learning, what concerns us more regarding a neural network is its depth as well as how depth impacts the behaviors of a network. Inspired by a width-depth symmetry consideration, we use a shortcut network to show that increasing the depth of a neural network can also give rise to a Gaussian process, which is a valuable addition to the existing theory and contributes to revealing the true picture of deep learning. Beyond the proposed Gaussian process by depth, we theoretically characterize its uniform tightness property and the smallest eigenvalue of the Gaussian process kernel. These characterizations can not only enhance our understanding of the proposed depth-induced Gaussian process but also pave the way for future applications. Lastly, we examine the performance of the proposed Gaussian process by regression experiments on two benchmark datasets.