This inventor holds 1 USPTO granted patent. Top assignee: Salesforce, 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: Kun Zhang and His Patent in Unsupervised Anomaly Detection.
Introduction
Kun Zhang is an inventor based in San Francisco, California. His patent work includes a system for detecting anomalies in datasets without relying on manually labeled examples.
Latest Patents
Zhang is associated with 1 patent, titled “Systems and Methods for Unsupervised Anomaly Detection.” The invention describes a system that receives a dataset containing anomalies and uses an inlier model to identify noisy labels. It trains a causality-based label-noise model using those labels and high-confidence instances, estimates the proportion of anomalies, and retrains the models iteratively until their outputs meet a convergence threshold. The system then extracts an anomaly-detection model from the trained label-noise model.
Career Highlights
Zhang works at Salesforce, Inc. His patent reflects an approach to improving anomaly detection by combining inlier estimation with causality-based modeling of label noise.
Collaborations
Zhang’s listed coworkers include Wenzhuo Yang and Chu Hong Hoi.
Conclusion
Kun Zhang’s patent describes an iterative, unsupervised approach to anomaly detection that aims to identify anomalies while accounting for noisy labels.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile