This inventor holds 1 USPTO granted patent. Top assignee: Corning Incorporated. 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: Zheren Wang and Data-Efficient Property Prediction
Introduction
Zheren Wang is an inventor based in Redwood City, California, and is associated with Corning Incorporated. His patent work addresses how machine-learning models can predict properties when only limited data is available.
Latest Patents
Wang is credited with 1 patent: “Collaborative Multitask and Transfer Learning for Predicting Properties with Scarce Data.” The invention describes training a shared encoder and distinct decoders on multiple property datasets, then using transfer learning to train a new decoder for a target property. The resulting model predicts a property that can guide the preparation of an item.
Career Highlights
Wang’s listed patent reflects an approach that combines collaborative multitask learning with transfer learning. By sharing an encoder across datasets while using separate decoders for different properties, the method is designed to make predictions useful even when data for a particular property is scarce.
Collaborations
The patent is associated with Corning Incorporated, connecting Wang’s invention to the company’s work in materials and technology. The patent description identifies a method for developing predictive models but does not specify particular collaborators.
Conclusion
Zheren Wang’s patent presents a data-efficient approach to predicting properties by combining shared learning and targeted transfer learning, with the goal of supporting the preparation of items based on predicted results.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
