Qiong Wu
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
Qiong Wu has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Jiajia Yang
Graduate School of Interdisciplinary Science and Engineering in Health Systems, Okayama University, Okayama 700-8530, Japan
2 shared publications - Yoshimichi Ejima
Graduate School of Interdisciplinary Science and Engineering in Health Systems, Okayama University, Okayama 700-8530, Japan
2 shared publications - Fengjie Zhang
School of Safety Science and Engineering, Henan Polytechnic University, Jiaozuo 454003, China
1 shared publication - Jie Meng
School of Computer Science and Technology, Soochow University, Suzhou, China
1 shared publication Jinglong Wu
The Graduate School of Natural Science and Technology, Okayama University, 3-1-1 Tsushima-naka, Japan
1 shared publication- Jinglong Wu
Graduate School of Interdisciplinary Science and Engineering in Health Systems, Okayama University, Okayama 700-8530, Japan
1 shared publication
Latest Publications
- 18 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. PP, 1-11
Fast Federated Learning Frameworks, Driven by Powerball Function and Hypergradient Descent
Federated learning (FL), a paradigm of distributed machine learning, aims to utilize user data for model training while safeguarding data privacy. However, numerous challenges arise in practical applications. Among these, accelerating the training rate and improving the generalization performance of FL models are particularly noteworthy challenges. Recent studies are particularly about understanding the role of stochastic optimization in federated optimization, leading to various fast federated optimization algorithms. As a special case, some studies have explored treating the sum of local differences as pseudo-gradients and also introduced the global learning rate. To deal with the challenge, this work develops fast federated optimization algorithms from the perspective of the Powerball function and the hypergradient descent, coined FedPBHD, extending the family of federated optimization algorithms. Different from existing FL algorithms, FedPBHD renews the update direction by using the discretization’s of ordinary differential equations (ODEs). Besides, FedPBHD utilizes an adaptive update rule of the learning rate (an extremely crucial hyperparameter for FL algorithms). The convergence performance of FedPBHD has been established through a rigorous theoretical analysis. Finally, we compare our FedPBHD algorithms against numerous benchmark optimizers for different practical applications of computer vision, natural language processing under both independent identically distribution (IID) and non-IID conditions. Experimental results on different problems have shown encouraging performance of the novel Federated optimization algorithms.
- 18 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. PP, 1-11
Fast Federated Learning Frameworks, Driven by Powerball Function and Hypergradient Descent
Federated learning (FL), a paradigm of distributed machine learning, aims to utilize user data for model training while safeguarding data privacy. However, numerous challenges arise in practical applications. Among these, accelerating the training rate and improving the generalization performance of FL models are particularly noteworthy challenges. Recent studies are particularly about understanding the role of stochastic optimization in federated optimization, leading to various fast federated optimization algorithms. As a special case, some studies have explored treating the sum of local differences as pseudo-gradients and also introduced the global learning rate. To deal with the challenge, this work develops fast federated optimization algorithms from the perspective of the Powerball function and the hypergradient descent, coined FedPBHD, extending the family of federated optimization algorithms. Different from existing FL algorithms, FedPBHD renews the update direction by using the discretization’s of ordinary differential equations (ODEs). Besides, FedPBHD utilizes an adaptive update rule of the learning rate (an extremely crucial hyperparameter for FL algorithms). The convergence performance of FedPBHD has been established through a rigorous theoretical analysis. Finally, we compare our FedPBHD algorithms against numerous benchmark optimizers for different practical applications of computer vision, natural language processing under both independent identically distribution (IID) and non-IID conditions. Experimental results on different problems have shown encouraging performance of the novel Federated optimization algorithms.
- 1 January 2022
- Published by Inderscience Publishers in International Journal of Mechatronics and Automation
- Vol. 1 (1) , 1
Investigating the difference of face distinction between adult and infant for design of service robots
- 1 January 2022
- Published by Inderscience Publishers in International Journal of Mechatronics and Automation
- Vol. 1 (1) , 1
Investigating the difference of face distinction between adult and infant for design of service robots
- 6 July 2019
- Published by Springer Nature
- p. 456-460
Interactive Design for Real-World Mapping of Virtual Reality
- 6 July 2019
- Published by Springer Nature
- p. 456-460
Interactive Design for Real-World Mapping of Virtual Reality
- 1 August 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 155, 1738-1743
A basic study on regular polygons recognition of central and peripheral vision field for virtual reality
Recently, virtual reality (VR) have received increasing attention. For the development of VR, human recognition mechanism is a crucial factor. Previous neuroimaging studies of the visual recognition have found that mechanism of recognition is different on retinal eccentricity of visual field. For example, when we recognize the face, the building and the scene, we use the central vision field. However not only the central vision field but also the peripheral vision field is used for building recognition. And for the recognize of the scene, the peripheral vision field is even more important than central vision field. Revealing the object recognition mechanism demands us to clarify the functions of the central and the peripheral vision field. The purpose of this study is the first step to reveal the object recognition mechanism of the central and the peripheral vision field. We analyze the dependency of the fundamental elements of the objects on retinal eccentricity. The retinal eccentricity is the degree of the gap between the central and the peripheral vision field. Basically, we measured the dependency of recognition on retinal eccentricity of visual field as we changed the elements and the exposure duration of stimuli. The fundamental elements of the objects are the stimuli size and the line width of the stimuli. The stimuli size and the line width of the stimuli showed a large dependency on retinal eccentricity of visual field. But the exposure duration of stimuli showed a small dependency on retinal eccentricity of visual field.
- 1 August 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 155, 1738-1743
A basic study on regular polygons recognition of central and peripheral vision field for virtual reality
Recently, virtual reality (VR) have received increasing attention. For the development of VR, human recognition mechanism is a crucial factor. Previous neuroimaging studies of the visual recognition have found that mechanism of recognition is different on retinal eccentricity of visual field. For example, when we recognize the face, the building and the scene, we use the central vision field. However not only the central vision field but also the peripheral vision field is used for building recognition. And for the recognize of the scene, the peripheral vision field is even more important than central vision field. Revealing the object recognition mechanism demands us to clarify the functions of the central and the peripheral vision field. The purpose of this study is the first step to reveal the object recognition mechanism of the central and the peripheral vision field. We analyze the dependency of the fundamental elements of the objects on retinal eccentricity. The retinal eccentricity is the degree of the gap between the central and the peripheral vision field. Basically, we measured the dependency of recognition on retinal eccentricity of visual field as we changed the elements and the exposure duration of stimuli. The fundamental elements of the objects are the stimuli size and the line width of the stimuli. The stimuli size and the line width of the stimuli showed a large dependency on retinal eccentricity of visual field. But the exposure duration of stimuli showed a small dependency on retinal eccentricity of visual field.