Dragan Zivkovic

Title: Dragan Zivkovic: Innovator in On-Device Machine Learning

Introduction

Dragan Zivkovic is a prominent inventor based in Sunnyvale, California, known for his significant contributions to the field of machine learning. With a total of 18 patents to his name, he has been at the forefront of developing innovative solutions that enhance the performance of on-device machine learning models.

Latest Patents

Dragan's latest patents focus on evaluating on-device machine learning models based on performance measures of client devices. One of his notable inventions involves systems and methods for assessing the performance of on-device ML models, which can include metrics such as latency, memory consumption, and CPU usage. These innovations allow client devices to process testing instances and generate performance measures, ensuring that the ML models are optimized for local use. Another significant patent involves using corrections of automated assistant functions for training on-device machine learning models. This invention enables client devices to utilize sensor data to activate dormant automated assistant functions based on predicted outputs, enhancing user interaction and functionality.

Career Highlights

Dragan Zivkovic has made remarkable strides in his career, particularly through his work at Google Inc. His expertise in machine learning has positioned him as a key player in the development of advanced technologies that improve user experience and device efficiency.

Collaborations

Throughout his career, Dragan has collaborated with notable colleagues, including Francoise Beaufays and Zhongli Ding. These partnerships have contributed to the successful development of innovative machine learning solutions.

Conclusion

Dragan Zivkovic's work in on-device machine learning exemplifies the impact of innovation in technology. His patents and contributions continue to shape the future of machine learning applications, making significant advancements in the field.