Company Filing History:
Years Active: 2022
Title: Innovations of Bodong Zhang in Deep Neural Networks
Introduction
Bodong Zhang is a prominent inventor based in Shenyang, China. He has made significant contributions to the field of deep neural networks, particularly in model compression techniques. With a total of 2 patents, his work focuses on enhancing the efficiency and performance of neural network models.
Latest Patents
Bodong Zhang's latest patents include "Lossless model compression by batch normalization layer pruning in deep neural networks." This patent proposes a method for pruning a batch normalization layer from a pre-trained deep neural network model. The process involves removing at least one batch normalization layer that is connected to a corresponding linear operation layer, such as a convolution layer or a dense layer. The weights of the corresponding linear operation layer are adjusted to compensate for the removal, resulting in a pruned candidate model that is utilized for inference.
Another notable patent is "Self-tuning incremental model compression solution in deep neural network with guaranteed accuracy performance." This method compresses a pre-trained deep neural network model by increasing its sparsity and quantizing the remaining weights into fixed-point representation. The accuracy of the compressed model is determined using an end-user training and validation data set, and the compression process is repeated until the accuracy meets or exceeds the desired performance metrics.
Career Highlights
Bodong Zhang is currently employed at Kneron Co., Ltd., a company based in Taiwan. His work at Kneron focuses on advancing deep learning technologies and improving model efficiency. His innovative approaches to model compression have positioned him as a key figure in the field.
Collaborations
Bodong Zhang collaborates with talented individuals such as Bike Xie and Junjie Su. Their combined expertise contributes to the development of cutting-edge solutions in deep neural networks.
Conclusion
Bodong Zhang's contributions to deep neural networks through his innovative patents demonstrate his commitment to advancing technology in this field. His work not only enhances model efficiency but also sets a foundation for future developments in artificial intelligence.