This inventor holds 1 USPTO granted patent and 1 published patent application. Top assignee: Meta Platforms Technologies, LLC. Active years: 2026.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2026
Title: Ziyun Li: Inventor in Sensor-Based Neural Network Systems.
Introduction
Ziyun Li is an inventor associated with Bellevue, Washington, and Meta Platforms Technologies, LLC. His patent work addresses how neural networks can be partitioned across sensor devices and an aggregator computing device.
Latest Patents
The provided information lists one patent for Li: “Network partitioning for sensor-based systems.” The patent describes methods, systems, and media for training multiple neural networks with different layer dimensions and compression-block locations. A network is selected according to the hardware constraints of the system, with one set of layers executed on sensor devices and another set executed on an aggregator computing device. Li’s listed patent portfolio includes 1 patent.
Career Highlights
Li’s identified patent reflects work on adapting neural-network architectures to distributed, sensor-based hardware. The approach considers where to place a compression block and how to divide network execution between sensors and an aggregator.
Collaborations
The supplied information names Xin Dong and Barbara De Salvo as Li’s coworkers. Their specific roles in the patent or related work are not provided.
Conclusion
Ziyun Li’s listed invention focuses on partitioning neural networks to accommodate hardware constraints in sensor-based systems, with the goal of distributing computation between sensor devices and an aggregator.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile