Sunnyvale, CA, United States of America

Sumeet Katariya

This inventor holds 2 USPTO granted patents. Top assignee: Amazon Technologies, Inc.. Active years: 2023.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Sumeet Katariya. Retrieved from https://idiyas.com/inventor/sumeet-katariya-sdjzzlgv

Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile


% Patents Active = 100.0

Average Co-Inventor Count = 4.4

ph-index = 1

Forward Citations = 3(Granted Patents)


Company Filing History:


Years Active: 2023

Loading Chart...
2 patents (USPTO):Explore Patents

Title: Sumeet Katariya: Innovator in Index Searching and Query Reformulation

Introduction

Sumeet Katariya is a notable inventor based in Sunnyvale, CA. He has made significant contributions to the fields of index searching and query reformulation. With a total of 2 patents, his work has the potential to enhance the efficiency of search technologies.

Latest Patents

Sumeet's latest patents include innovative technologies that address complex challenges in search queries. One of his patents focuses on "Index searching with non-Euclidean representations." This invention allows items within an index to be converted from classic geometry and embedded into a hyperbolic space. The hyperboloids in this hyperbolic space enable higher precision classifications of items relative to their hierarchical structure. Additionally, a received search query can be transformed into hyperbolic space and mapped as a query hyperboloid against an answer space containing hyperboloids for items within the index. This method allows for the determination of distances or overlaps between hyperboloids to generate a set of search results.

Another significant patent is "Reformulation of tail queries for product searches." This technology provides a method for reformulating a tail query into a head query with the same purchase intent by mapping the tail query to the head query. The invention includes learning a reasonable embedding on historical head queries, which can be refined using rewards generated from a persistently noisy oracle. This approach compensates for the lack of historical behavioral signals for tail queries. Furthermore, a contextual sampling technique utilizing text-based or oracle-based rewards can be implemented to mitigate biases introduced by persistent noise in the oracle. Numerical experiments on large-scale e-commerce datasets demonstrate that these technologies can outperform several conventional approaches to query reformulation.

Career Highlights

Sumeet Katariya is currently employed at Amazon Technologies, Inc., where he continues to innovate and develop advanced search technologies. His work has garnered attention for its practical applications in enhancing user experience in e-commerce.

Collaborations

Sumeet collaborates with talented individuals such as Nikhil S Rao and Karthik Subbian, contributing to a dynamic and innovative work environment.

Conclusion

Sumeet Katariya's contributions to index searching and query reformulation showcase his expertise and commitment to advancing search technologies. His patents reflect a deep understanding of complex algorithms and their applications in real-world scenarios.

Profile summary based on public USPTO records.
Data Sources: USPTO Patent Grant XML, Patent Center, EPO & CIPO • Normalized by IDiyas Innovation Graph. Methodology & provenance architecturePlease report any incorrect information to [email protected]
Loading…