Liran Levy

Title: Liran Levy: Innovator in Neural Network Processing

Introduction

Liran Levy is a prominent inventor based in Kiryat Ata, Israel. He has made significant contributions to the field of neural network processing, holding 2 patents that showcase his innovative approach to technology.

Latest Patents

One of Liran's latest patents is focused on a neural network processor. This method involves applying a group of neural network kernels from respective layers of a neural network to generate output results. The process includes applying a 2D kernel on 2D layer input values to produce first layer output values. Additionally, it involves scanning the 2D layer input values with the 2D kernel, which includes flattening both the kernel and the input values, and storing the output in a 3D memory unit. Another significant patent addresses transposed convolution on downsampled data. This invention provides solutions to technical challenges related to preserving the spatial dimensions of images used in Convolutional Neural Networks (CNNs). By utilizing transposed convolutional layers, Liran's method enhances spatial dimension preservation, particularly for Advanced Driver Assistance Systems (ADAS) and Autonomous Vehicles (AV).

Career Highlights

Liran Levy is currently employed at Mobileye Vision Technologies Ltd., a company renowned for its advancements in autonomous driving technologies. His work at Mobileye has positioned him at the forefront of innovation in the field of computer vision and neural networks.

Collaborations

Liran collaborates with talented individuals such as Ariel Binenfeld and Yaniv Fais, contributing to a dynamic team focused on pushing the boundaries of technology.

Conclusion

Liran Levy's contributions to neural network processing and his innovative patents reflect his commitment to advancing technology in the field of computer vision. His work continues to influence the development of systems that enhance safety and efficiency in autonomous driving applications.