This inventor holds 1 USPTO granted patent. Top assignee: Samsung Electronics Co., Ltd.. Active years: 2021.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2021
Title: Oded Ilan-Lipowsky: Innovator in Convolution Neural Network Computation
Introduction
Oded Ilan-Lipowsky is a notable inventor based in Tel Aviv, Israel. He has made significant contributions to the field of hardware acceleration for convolutional neural networks. His innovative approach has led to the development of a unique method that enhances computational efficiency.
Latest Patents
Oded holds a patent for a "Low-power hardware acceleration method and system for convolution neural network computation." This patent describes a method of processing a layer in a convolutional neural network. It involves reading various subsets of pairs of input vector values and corresponding weight vector values from on-chip memory. The method allows for simultaneous processing of these pairs in multiplication units, ultimately producing an output vector for the current layer. The technique utilizes floating point values and incorporates a log domain for efficient computation.
Career Highlights
Oded is currently employed at Samsung Electronics Co., Ltd., where he continues to push the boundaries of technology. His work focuses on improving the performance of neural networks, which are crucial for various applications in artificial intelligence and machine learning.
Collaborations
Oded collaborates with talented individuals such as Shai Litvak and Evgeny Soloveichik. Their combined expertise fosters an environment of innovation and creativity, leading to groundbreaking advancements in their field.
Conclusion
Oded Ilan-Lipowsky is a prominent figure in the realm of neural network computation, with a patent that showcases his innovative spirit. His contributions to technology are paving the way for more efficient and powerful computational methods.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
