Intellectual Property India Publishes Patent Application for 'Federated Learning Architecture With Differential Privacy Guarantees For Secure Collaborative Training' Filed by Vardhaman College Of Engineering

MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541123674 A) filed by Vardhaman College Of Engineering, Hyderabad, Telangana, on Dec. 8, 2025, for 'federated learning architecture with differential privacy guarantees for secure collaborative training.'

Inventor(s) include Mr. Nirmal Keshari Swain; Mr. M Yugandhar; Ms. B Swapna; Dr. K Nikhila; Ms. Farhana Begum; and Dr. L. Sunitha.

The application for the patent was published on Jan. 2, under issue no. 01/2026.

According to the abstract released by the Intellectual Property India: "The invention provides a federated learning architecture that enables multiple distributed clients to collaboratively train a global machine learning model while ensuring data privacy through differential privacy guarantees. Each client computes local model updates from private datasets, applies calibrated noise to protect individual data points, and transmits encrypted or masked updates to a central aggregator. The aggregator combines updates using an adaptive weighting mechanism that accounts for data quality, distribution, and client participation, producing a global model without accessing raw client data. A privacy controller tracks cumulative privacy loss across training rounds, and a communication management module optimizes update transmission to reduce network overhead. The system ensures secure, efficient, and privacy-compliant collaborative training suitable for sensitive domains such as healthcare, finance, IoT, and governmental applications."

Disclaimer: Curated by HT Syndication.