Chang Wan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Chang Wan has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Li Qiang Zhu
Ningbo University
1 shared publication - Ming-Hsuan Yang
University of California, Merced, USA
1 shared publication - Minglu Li
Zhejiang Normal University
1 shared publication - Ping Feng
RUDN University
1 shared publication - Qing Wan
YongJiang Laboratory
1 shared publication - Wei Wang
National University of Defense Technology
1 shared publication
Latest Publications
- Open Access
- 10 January 2025
- Published by Association for Computing Machinery (ACM)
- p. 699-705
Continuous Iterative Image Editing Based on Diffusion Models
- 10 January 2025
- Published by Association for Computing Machinery (ACM)
- p. 699-705
Open AccessContinuous Iterative Image Editing Based on Diffusion Models
- 11 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 35 (1) , 670-683
Nested Annealed Training Scheme for Generative Adversarial Networks
Recently, researchers have proposed many deep generative models, including generative adversarial networks (GANs) and denoising diffusion models. Although significant breakthroughs have been made and empirical success has been achieved with the GAN, its mathematical underpinnings remain relatively unknown. This paper focuses on a rigorous mathematical theoretical framework: the composite-functional-gradient GAN (CFG) [1]. Specifically, we reveal the theoretical connection between the CFG model and score-based models. We find that the CFG discriminator’s training objective is equivalent to finding an optimal D (x). The optimal D (x)’s gradient differentiates the integral of the differences between the score functions of real and synthesized samples. Conversely, training the CFG generator involves finding an optimal G (x) that minimizes this difference. In this paper, we aim to derive an annealed weight preceding the CFG discriminator’s weight. This new explicit theoretical explanation model is called the annealed CFG method. To overcome the annealed CFG method’s limitation, as the method is not readily applicable to the state-of-the-art (SOTA) GAN model, we propose a nested annealed training scheme (NATS). This scheme keeps the annealed weight from the CFG method and can be seamlessly adapted to various GAN models, no matter their structural, loss, or regularization differences. We conduct thorough experimental evaluations on various benchmark datasets for image generation. The results show that our annealed CFG and NATS methods significantly improve the synthesized samples’ quality and diversity. This improvement is clear when comparing the CFG method and the SOTA GAN models.
- 11 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 35 (1) , 670-683
Nested Annealed Training Scheme for Generative Adversarial Networks
Recently, researchers have proposed many deep generative models, including generative adversarial networks (GANs) and denoising diffusion models. Although significant breakthroughs have been made and empirical success has been achieved with the GAN, its mathematical underpinnings remain relatively unknown. This paper focuses on a rigorous mathematical theoretical framework: the composite-functional-gradient GAN (CFG) [1]. Specifically, we reveal the theoretical connection between the CFG model and score-based models. We find that the CFG discriminator’s training objective is equivalent to finding an optimal D (x). The optimal D (x)’s gradient differentiates the integral of the differences between the score functions of real and synthesized samples. Conversely, training the CFG generator involves finding an optimal G (x) that minimizes this difference. In this paper, we aim to derive an annealed weight preceding the CFG discriminator’s weight. This new explicit theoretical explanation model is called the annealed CFG method. To overcome the annealed CFG method’s limitation, as the method is not readily applicable to the state-of-the-art (SOTA) GAN model, we propose a nested annealed training scheme (NATS). This scheme keeps the annealed weight from the CFG method and can be seamlessly adapted to various GAN models, no matter their structural, loss, or regularization differences. We conduct thorough experimental evaluations on various benchmark datasets for image generation. The results show that our annealed CFG and NATS methods significantly improve the synthesized samples’ quality and diversity. This improvement is clear when comparing the CFG method and the SOTA GAN models.
- 9 May 2016
- Published by Wiley in Advanced Materials
- Vol. 28 (28) , 5878-5885
Flexible Metal Oxide/Graphene Oxide Hybrid Neuromorphic Transistors on Flexible Conducting Graphene Substrates
Flexible metal oxide/graphene oxide hybrid multi‐gate neuromorphic transistors are fabricated on flexible conducting graphene substrates. Dendritic integrations in both spatial and temporal modes are emulated, and spatiotemporal correlated logics are obtained. A proof‐of‐principle visual system model for emulating Lobula Giant Motion Detector neuron is also investigated. The results are of great significance for flexible sensors and neuromorphic cognitive systems.
- 9 May 2016
- Published by Wiley in Advanced Materials
- Vol. 28 (28) , 5878-5885
Flexible Metal Oxide/Graphene Oxide Hybrid Neuromorphic Transistors on Flexible Conducting Graphene Substrates
Flexible metal oxide/graphene oxide hybrid multi‐gate neuromorphic transistors are fabricated on flexible conducting graphene substrates. Dendritic integrations in both spatial and temporal modes are emulated, and spatiotemporal correlated logics are obtained. A proof‐of‐principle visual system model for emulating Lobula Giant Motion Detector neuron is also investigated. The results are of great significance for flexible sensors and neuromorphic cognitive systems.