Zeyuan Chen

Title: Zeyuan Chen: Innovator in Online Adaptation and Field Extraction Technologies

Introduction

Zeyuan Chen is a prominent inventor based in Mountain View, CA. He has made significant contributions to the fields of online adaptation and field extraction technologies. With a total of 8 patents, his work has advanced the capabilities of data processing and machine learning.

Latest Patents

One of Zeyuan Chen's latest patents is titled "Systems and methods for online adaptation for cross-domain streaming data." This invention provides an online domain adaptation framework that utilizes cross-domain bootstrapping. The framework allows for the immediate deletion of target domain streaming data after adaptation. By increasing data diversity across domains, it trains independent learners to preserve differences and co-supervise learning on the target domain.

Another notable patent is "Systems and methods for field extraction from unlabeled data." This system eliminates the need for field-level annotations during training. It employs a transformer-based structure to model interactions between text tokens and predict field tags. A refinement module is included to enhance the accuracy of pseudo-labels, ensuring that the training process is robust against noise.

Career Highlights

Zeyuan Chen has worked with notable companies such as Salesforce, Inc. and Salesforce.com, Inc. His experience in these organizations has contributed to his expertise in developing innovative solutions in data processing and machine learning.

Collaborations

Throughout his career, Zeyuan has collaborated with talented individuals, including Ran Xu and Mingfei Gao. These partnerships have fostered a creative environment that has led to the development of groundbreaking technologies.

Conclusion

Zeyuan Chen is a distinguished inventor whose work in online adaptation and field extraction has made a significant impact in the tech industry. His innovative patents and collaborations highlight his commitment to advancing technology and improving data processing methodologies.