Brian Kahne

Title: Brian Kahne: Innovator in Neural Network Quantization

Introduction

Brian Kahne is an accomplished inventor based in Austin, TX. He has made significant contributions to the field of neural networks, particularly in the area of quantization techniques. His innovative approach has the potential to enhance the efficiency of neural network processing.

Latest Patents

Brian Kahne holds a patent for "Simulated low bit-width quantization using bit shifted neural network parameters." This processor-implemented method involves bit shifting a binary representation of a neural network parameter. The neural network parameter has fewer bits, b, than the number of hardware bits, B, supported by the hardware that processes the neural network parameter. The bit shifting effectively multiplies the neural network parameter by 2. The method also includes dividing a quantization scale by 2 to obtain an updated quantization scale. Furthermore, the method quantizes the bit shifted binary representation with the updated quantization scale to obtain a value for the neural network parameter. This patent showcases his innovative thinking and technical expertise.

Career Highlights

Brian Kahne is currently employed at Qualcomm Incorporated, a leading technology company known for its advancements in telecommunications and semiconductor technology. His work at Qualcomm has allowed him to apply his knowledge and skills in a practical setting, contributing to the development of cutting-edge technologies.

Collaborations

Throughout his career, Brian has collaborated with notable colleagues, including Marinus Willem Van Baalen and Eric W Mahurin. These collaborations have fostered a creative environment that encourages innovation and the sharing of ideas.

Conclusion

Brian Kahne is a prominent inventor whose work in neural network quantization is paving the way for more efficient processing methods. His contributions to the field are significant and reflect his dedication to advancing technology.