This inventor holds 1 USPTO granted patent. Top assignee: Leidos, 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: Manbir S. Gulati and Synthetic Data Generation
Introduction
Manbir S. Gulati is an inventor based in Frederick, Maryland, United States. His patent work includes a transformer-based system for generating synthetic data from datasets containing multiple fields.
Latest Patents
Gulati is associated with 1 patent, titled “System and Method for Generating Synthetic Data with Masked Transformers.” The invention describes a modeling architecture that embeds input data and constructs separate embedding matrices for its fields. A masking model replaces selected real data values with mask tokens, and a transformer predicts the masked values over multiple iterations. This approach is designed to train a generator to produce synthetic data.
Career Highlights
Gulati is associated with Leidos, Inc. His patent reflects work on applying transformer models to synthetic data generation.
Collaborations
Gulati’s listed coworker is Paul F. Roysdon. The available information identifies their professional association but does not specify their respective contributions to the patent.
Conclusion
Manbir S. Gulati’s patent presents a transformer-based method for generating synthetic data by iteratively predicting masked values, highlighting his work in data modeling and artificial intelligence.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile