Jiaxin Shi

Title: Innovations of Jiaxin Shi in Neural Network Quantization

Introduction

Jiaxin Shi is an accomplished inventor based in Shanghai, China. He has made significant contributions to the field of neural networks, particularly in the area of quantization techniques. His work focuses on improving the efficiency of neural network inference processes.

Latest Patents

Jiaxin Shi holds a patent for "Asymmetric quantization for compression and for acceleration of inference for neural networks." This patent presents embodiments of an improved asymmetric quantization, referred to as improved asymmetric quantization (IAQ). The IAQ embodiments combine the benefits of conventional asymmetric quantization and symmetric quantization while providing additional computational efficiencies. By adopting an asymmetric range of the weights of a neural network layer, IAQ circumvents the limitations of symmetric quantization. Furthermore, the inference process of a neural network quantized by IAQ is significantly faster than that of conventional asymmetric quantization, as it quantizes an offset value of each layer.

Career Highlights

Jiaxin Shi is currently employed at Baidu USA LLC, where he continues to innovate in the field of artificial intelligence and machine learning. His work has garnered attention for its potential to enhance the performance of neural networks in various applications.

Collaborations

Jiaxin has collaborated with notable colleagues, including Yingzhen Yang and Zhibiao Zhao, contributing to advancements in their shared field of expertise.

Conclusion

Jiaxin Shi's innovative work in asymmetric quantization represents a significant advancement in neural network technology. His contributions are poised to impact the efficiency and speed of machine learning applications in the future.