Gilad Nahor

Title: Innovations by Gilad Nahor in Artificial Neural Networks

Introduction

Gilad Nahor is an accomplished inventor based in Modi'in-Maccabim-Re'ut, Israel. He has made significant contributions to the field of artificial intelligence, particularly in the development of artificial neural networks. With a total of two patents to his name, Nahor's work focuses on improving power performance and memory efficiency in neural network processing.

Latest Patents

Nahor's latest patents include "Structured weight based sparsity in an artificial neural network" and "Structured activation based sparsity in an artificial neural network." Both inventions present novel systems and methods that enhance power performance while reducing memory requirements for artificial neural networks. These innovations utilize structured sparsity mechanisms to optimize memory packing, resulting in considerably reduced memory usage. The sparsity guided training mechanism synthesizes structured sparsity weights, while a compiler mechanism within a software development kit (SDK) manipulates these weights to generate a sparse set of static weights for the neural network. The application of structured sparsity not only lowers the span of search options but also creates a relatively loose coupling between the data and control planes.

Career Highlights

Gilad Nahor is currently employed at Hailo Technologies Ltd., where he continues to push the boundaries of artificial intelligence technology. His work has been instrumental in advancing the capabilities of neural network processing engines.

Collaborations

Nahor collaborates with talented individuals such as Avi Baum and Or Danon, contributing to a dynamic and innovative work environment.

Conclusion

Gilad Nahor's contributions to artificial neural networks through his patents demonstrate his commitment to innovation in the field of artificial intelligence. His work not only enhances the efficiency of neural networks but also paves the way for future advancements in technology.