Yanjun Zhang
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
Research Subjects
Yanjun Zhang has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Guangdong Bai
City University of Hong Kong, Hong Kong, China
2 shared publications - Ryan Kok Leong Ko
University of Queensland
2 shared publications - Xue Li
The University of Queensland
2 shared publications - Caitlin Curtis
Business School The University of Queensland Brisbane Queensland Australia
1 shared publication - Chen Chen
School of Biomedical Sciences, The University of Queensland, Brisbane, QLD 4072, Australia
1 shared publication - Jingpeng Li
Inspur Software Group Co., Ltd., Data Center Service Business Dept, Jinan 250000, China
1 shared publication
Latest Publications
- Open Access
- 14 February 2022
- Published by Wiley in Scientific Programming
- Vol. 2022, 1-11
An Intelligent Vehicle Alarm User Terminal System Based on Emotional Identification Technology
Negative emotions could increase the risks of traffic accidents. However, the driver’s emotional identification is rarely considered in the current design of intelligent vehicle alarm user terminals (IVAUTs). To solve the problem, this paper tries to design an IVAUT system based on emotional identification technology. Firstly, the transformer network was combined with a convolutional neural network (CNN) into a voice emotional identification system for intelligent vehicle alarm, and an emotional labeling approach was provided. Next, a bimodal fusion model was developed based on decompose-CNNs, which includes an appearance module, an optical flow module, and a bimodal fusion module. The proposed emotional identification approach was proved effective through experiments.
- 14 February 2022
- Published by Wiley in Scientific Programming
- Vol. 2022, 1-11
Open AccessAn Intelligent Vehicle Alarm User Terminal System Based on Emotional Identification Technology
Negative emotions could increase the risks of traffic accidents. However, the driver’s emotional identification is rarely considered in the current design of intelligent vehicle alarm user terminals (IVAUTs). To solve the problem, this paper tries to design an IVAUT system based on emotional identification technology. Firstly, the transformer network was combined with a convolutional neural network (CNN) into a voice emotional identification system for intelligent vehicle alarm, and an emotional labeling approach was provided. Next, a bimodal fusion model was developed based on decompose-CNNs, which includes an appearance module, an optical flow module, and a bimodal fusion module. The proposed emotional identification approach was proved effective through experiments.
- Open Access
Confined Gradient Descent: Privacy-preserving Optimization for Federated Learning
Federated learning enables multiple participants to collaboratively train a model without aggregating the training data. Although the training data are kept within each participant and the local gradients can be securely synthesized, recent studies have shown that such privacy protection is insufficient. The global model parameters that have to be shared for optimization are susceptible to leak information about training data. In this work, we propose Confined Gradient Descent (CGD) that enhances privacy of federated learning by eliminating the sharing of global model parameters. CGD exploits the fact that a gradient descent optimization can start with a set of discrete points and converges to another set at the neighborhood of the global minimum of the objective function. It lets the participants independently train on their local data, and securely share the sum of local gradients to benefit each other. We formally demonstrate CGD's privacy enhancement over traditional FL. We prove that less information is exposed in CGD compared to that of traditional FL. CGD also guarantees desired model accuracy. We theoretically establish a convergence rate for CGD. We prove that the loss of the proprietary models learned for each participant against a model learned by aggregated training data is bounded. Extensive experimental results on two real-world datasets demonstrate the performance of CGD is comparable with the centralized learning, with marginal differences on validation loss (mostly within 0.05) and accuracy (mostly within 1%).
Open AccessConfined Gradient Descent: Privacy-preserving Optimization for Federated Learning
Federated learning enables multiple participants to collaboratively train a model without aggregating the training data. Although the training data are kept within each participant and the local gradients can be securely synthesized, recent studies have shown that such privacy protection is insufficient. The global model parameters that have to be shared for optimization are susceptible to leak information about training data. In this work, we propose Confined Gradient Descent (CGD) that enhances privacy of federated learning by eliminating the sharing of global model parameters. CGD exploits the fact that a gradient descent optimization can start with a set of discrete points and converges to another set at the neighborhood of the global minimum of the objective function. It lets the participants independently train on their local data, and securely share the sum of local gradients to benefit each other. We formally demonstrate CGD's privacy enhancement over traditional FL. We prove that less information is exposed in CGD compared to that of traditional FL. CGD also guarantees desired model accuracy. We theoretically establish a convergence rate for CGD. We prove that the loss of the proprietary models learned for each participant against a model learned by aggregated training data is bounded. Extensive experimental results on two real-world datasets demonstrate the performance of CGD is comparable with the centralized learning, with marginal differences on validation loss (mostly within 0.05) and accuracy (mostly within 1%).
- 12 September 2020
- Published by Springer Nature
- p. 399-418
PrivColl: Practical Privacy-Preserving Collaborative Machine Learning
- 12 September 2020
- Published by Springer Nature
- p. 399-418
PrivColl: Practical Privacy-Preserving Collaborative Machine Learning