Deep neural network oriented evolutionary parametric eye modeling

Zheng, Yang; Hong, Fei; Li, Ruimin; Hsung, Tai-Chiu; Song, Zongxi; Wen, Desheng

doi:10.1016/j.patcog.2020.107755

Cited by 5 publications

(2 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The BP neural network has outstanding merits and can effectively solve the problem of adjusting the weights and thresholds of multilayer feed-forward neural networks. It features massively parallel processing, strong fault tolerance, and distributed storage [42]. With a complete theoretical foundation and successful application cases, it implements a nonlinear mapping function from the input layer to the output layer and has become a hot topic for applications in several research areas.…”

Section: Ga-pso-bp Neural Networkmentioning

confidence: 99%

Path Optimization of Technological Innovation Efficiency Improvement in China’s High-Tech Industries Based on QCA and GA-PSO-BP Neural Network

Kou

Liu

Zhou

et al. 2023

Systems

View full text Add to dashboard Cite

Innovation is the main driving force to promote national technological progress. It is of great significance to explore the optimal path to improve innovation efficiency by using the qualitative method and neural network prediction model to promote the high-quality development of the national economy. This study focuses on high-tech industries in the eastern, central and western regions of China; a factor-dependent research framework for innovation efficiency improvement in high-tech industries is constructed in China. The fuzzy-set qualitative comparative analysis method (QCA) is used to explore multiple paths to enhance the innovation efficiency of China’s high-tech industries. Then, a GA-PSO-BP neural network is used to construct an optimization model for the enhancement path of technological innovation efficiency, which clarifies the optimal path for the enhancement of innovation efficiency of high-tech industries in the eastern, central and western regions of China. Finally, innovation management strategies for high-tech industries are presented with regional features. The study finds that none of the individual conditions are necessary to promote the innovation efficiency of China’s high-tech industries, and only the linkage effect of the factors can achieve the goal of improving the innovation efficiency level of China’s high-tech industries. There are four configuration paths to improve the innovation efficiency of China’s high-tech industries, which are: “Multinational company (MNC) innovation—economic development—government support”; “MNC innovation—government support”; “economic development—government support”; and “economic development”. The characteristics of regional heterogeneity make differences in the optimal paths of innovation efficiency improvement in high-tech industries in eastern, central and western regions of China.

show abstract

Section: Ga-pso-bp Neural Networkmentioning

confidence: 99%

Path Optimization of Technological Innovation Efficiency Improvement in China’s High-Tech Industries Based on QCA and GA-PSO-BP Neural Network

Kou

Liu

Zhou

et al. 2023

Systems

View full text Add to dashboard Cite

show abstract

“…Inspired by biological evolution, evolutionary algorithms have succeeded in many computing tasks, including optimization, modeling, and design [29] , [30] . Evolutionary algorithms often perform well-approximating solutions to almost all types of problems with high robustness.…”

Section: Related Workmentioning

confidence: 99%

An evolvable adversarial network with gradient penalty for COVID-19 infection segmentation

Zhu

Zhang

et al. 2021

Applied Soft Computing

View full text Add to dashboard Cite

COVID-19 infection segmentation has essential applications in determining the severity of a COVID-19 patient and can provide a necessary basis for doctors to adopt a treatment scheme. However, in clinical applications, infection segmentation is performed by human beings, which is time-consuming and generally introduces bias. In this paper, we developed a novel evolvable adversarial framework for COVID-19 infection segmentation. Three generator networks compose an evolutionary population to accommodate the current discriminator, i.e., generator networks evolved with different mutations instead of the single adversarial objective to provide sufficient gradient feedback Compared with the existing work that enforces a Lipschitz constraint by weight clipping, which may lead to gradient exploding or vanishing, the proposed model also incorporates the gradient penalty into the network, penalizing the discriminator’s gradient norm input. Experiments on several COVID-19 CT scan datasets verified that the proposed method achieved superior effectiveness and stability for COVID-19 infection segmentation.

show abstract

Attention-Aware Recognition of Activities of Daily Living Based on Eye Gaze Tracking

Madhusanka

Ramadass

Rajagopal

et al. 2022

Internet of Things for Human-Centered Design

View full text Add to dashboard Cite

Deep neural network oriented evolutionary parametric eye modeling

Cited by 5 publications

References 31 publications

Path Optimization of Technological Innovation Efficiency Improvement in China’s High-Tech Industries Based on QCA and GA-PSO-BP Neural Network

Path Optimization of Technological Innovation Efficiency Improvement in China’s High-Tech Industries Based on QCA and GA-PSO-BP Neural Network

An evolvable adversarial network with gradient penalty for COVID-19 infection segmentation

Attention-Aware Recognition of Activities of Daily Living Based on Eye Gaze Tracking

Contact Info

Product

Resources

About