Han Li

Title: Han Li - Innovator in User Embedding Systems

Introduction

Han Li is a prominent inventor based in Sunnyvale, CA (US). He has made significant contributions to the field of recommendation systems through his innovative patent. His work focuses on enhancing user experience by utilizing advanced user embeddings.

Latest Patents

Han Li holds a patent titled "Recommendation method, product, and system using user embeddings having stable long-term component and dynamic short-term session component." This patent describes an online system that employs a two-tower model to identify candidate items for presentation to users. The model consists of an item tower that computes item embeddings and a user tower that computes user embeddings. The user tower is further divided into a long-term sub-tower for stable user embeddings and a short-term sub-tower for dynamic session-based embeddings. This innovative approach is trained using item data, user data, and session data, allowing for a more personalized user experience.

Career Highlights

Han Li is currently employed at Maplebear Inc., where he continues to develop cutting-edge technologies in the field of recommendation systems. His expertise in user embeddings has positioned him as a valuable asset to his company and the industry.

Collaborations

Some of Han Li's notable coworkers include David Vengerov and Chuanwei Ruan. Their collaborative efforts contribute to the advancement of technology within their organization.

Conclusion

Han Li's innovative work in user embedding systems showcases his commitment to improving user experiences through technology. His patent reflects a significant advancement in recommendation methods, highlighting his role as a leading inventor in this field.