Zon-yin Shae

Title: Zon-yin Shae and Multi-Layer Federated Learning

Introduction

Zon-yin Shae is an inventor associated with Asia University in Taipei, Taiwan. Her work includes a patent on distributed clustering for multi-layer federated learning.

Latest Patents

Shae’s patent, “Computer Implemented Multi-Layer Federated Learning Method Based on Distributed Clustering and Non-Transitory Computer Readable Medium Thereof,” describes a method for grouping participating nodes according to the similarity of their feature distributions. The nodes’ local models are updated using federated learning, then processed through a multi-layer aggregation mechanism until the clustering results meet specified requirements.

The patent also describes a blockchain-based system built around this distributed-clustering architecture. Its components include a model aggregation module, an API module, a time-series synchronization module, and IPFS. The approach is intended to improve learning performance across non-independent and identically distributed data sets. Shae is associated with 1 patent.

Career Highlights

Shae’s patent work addresses challenges in federated learning, where data remains distributed across participating nodes. By combining clustering with layered model aggregation, the disclosed method aims to account for differences in data distributions while coordinating model updates.

Collaborations

The supplied information identifies Kun-yi Chen and Jing-pha Tsai as Shae’s coworkers. They are associated with her professional context at Asia University.

Conclusion

Zon-yin Shae’s patented work focuses on improving federated learning through distributed clustering, layered aggregation, and blockchain-based system components.