This inventor holds 1 USPTO granted patent. Top assignee: Ansys, Inc.. 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: Lars Gräning and Neural Network Architecture Design.
Introduction
Lars Gräning is an inventor associated with Munich, Germany, and Ansys, Inc. His patent work concerns methods for designing artificial neural network architectures.
Latest Patents
Gräning is credited with 1 patent: “Automated Design of Architectures of Artificial Neural Networks.” The invention describes a method and apparatus for determining a reduced-space neural network architecture. A device receives a full-space architecture containing nodes and weights, transforms the weights, and uses them to reduce the nodes. It then creates a reduced-space architecture from the resulting nodes.
Career Highlights
Gräning’s listed invention addresses the design of neural network architectures by reducing their structure through a process informed by transformed weights. He is associated with Ansys, Inc.
Collaborations
The provided information lists Anas Abdulhkim Aburawwash Abdulhkim and Thomas Most as Gräning’s coworkers.
Conclusion
Lars Gräning’s patent reflects work on streamlining artificial neural network architecture design through weight-guided node reduction.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
