MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202641061375 A) filed by Dr. A. Muthusamy; Dr. M. Kannan; Dr. M. Vijayakumar; Dr. Savithri M; Dr. D. Maheshwari; and Dr. J. Vaishnavi, Erode, Tamil Nadu, on May 14, for 'system and method for federated learning-based privacy-preserving artificial intelligence.'
Inventor(s) include Dr. A. Muthusamy; Dr. M. Kannan; Dr. M. Vijayakumar; Dr. Savithri M; Dr. D. Maheshwari; and Dr. J. Vaishnavi.
The application for the patent was published on May 29, under issue no. 22/2026.
According to the abstract released by the Intellectual Property India: "The invention discloses a federated learning-based system for privacy-preserving artificial intelligence. The system enables collaborative model training across distributed client nodes without requiring centralized data aggregation. Each client performs local training on private datasets and transmits encrypted updates to a central server, which employs secure aggregation protocols. Differential privacy mechanisms ensure that individual contributions cannot be reverse-engineered, while adaptive communication strategies optimize bandwidth usage. Anomaly detection modules safeguard against adversarial attacks, and compliance blocks provide transparency and regulatory adherence. The invention enhances privacy, scalability, and trustworthiness in AI applications across sensitive domains such as healthcare, finance, and personalized services."
Disclaimer: Curated by HT Syndication.
