MUMBAI, India, Feb. 6 -- Intellectual Property India has published a patent application (202511132710 A) filed by Dr. Akash Kumar; Anjali Singh; Sonali Gahlot; Ms. Harnit Saini; Rajeev Gupta; Dr. Amrita Bhatnagar; and Pronab Kumar Adhikari, Ghaziabad, Uttar Pradesh, on Dec. 27, 2025, for 'a system and method for federated artificial intelligence-based privacy-preserving learning in smart devices.'
Inventor(s) include Dr. Akash Kumar; Anjali Singh; Sonali Gahlot; Ms. Harnit Saini; Rajeev Gupta; Dr. Amrita Bhatnagar; and Pronab Kumar Adhikari.
The application for the patent was published on Feb. 6, under issue no. 06/2026.
According to the abstract released by the Intellectual Property India: "The present invention provides a system and method for federated artificial intelligence-based privacy-preserving learning across a plurality of smart devices. Each device locally trains a machine learning model using private data and transmits only encrypted model updates to a federated coordination entity for secure aggregation. The system incorporates privacy-preserving techniques such as differential privacy, homomorphic encryption, and secure multiparty computation to prevent exposure of sensitive information. Adaptive device participation, fault-tolerant aggregation, and anomaly detection mechanisms ensure robustness, scalability, and security across heterogeneous smart device networks. The invention is applicable to healthcare, smart homes, industrial IoT, autonomous transportation, and other intelligent distributed environments, providing efficient, privacy-compliant, and collaborative AI model training."
Disclaimer: Curated by HT Syndication.
