Company Filing History:
Years Active: 2023-2025
Title: Zhixin Pan: Innovator in Machine Learning and Hardware Security
Introduction
Zhixin Pan is a prominent inventor based in Gainesville, FL (US), known for his contributions to the fields of machine learning and hardware security. With a total of 4 patents, Pan has made significant strides in developing innovative solutions that address vulnerabilities in detection systems.
Latest Patents
One of Pan's latest patents focuses on AI-based trojans for evading machine learning detection. This invention provides a robust backdoor attack on machine learning (ML)-based detection systems, demonstrating and identifying their vulnerabilities. The AI-based Trojan attack is generated and implanted within an ML model trained for classification and detection tasks. It can be triggered by specific inputs to manipulate the expected outputs of the ML model. This analysis enables the identification of vulnerabilities and the design of more secure ML models. Another notable patent involves hardware Trojan detection using reinforcement learning. This method includes generating test patterns to activate hardware Trojans within integrated circuit designs. By evaluating nodes of the circuit and assigning rareness and testability values, the reinforcement learning model identifies optimal test patterns for triggering nodes effectively.
Career Highlights
Zhixin Pan is affiliated with the University of Florida Research Foundation, Incorporated, where he continues to advance research in his areas of expertise. His work has garnered attention for its innovative approach to enhancing security in machine learning and hardware systems.
Collaborations
Pan collaborates with notable colleagues, including Prabhat Kumar Mishra and Jennifer Marie Sheldon, who contribute to his research endeavors.
Conclusion
Zhixin Pan's innovative work in machine learning and hardware security showcases his commitment to advancing technology and addressing critical vulnerabilities. His contributions are paving the way for more secure systems in the future.