MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641049999 A) filed by Sr University, Warangal, Telangana, on April 20, for 'a federated bayesian learning framework for privacy-preserving distributed forecasting.'
Inventor(s) include Dr. Sanju; Dr. Suresh Yarlagadda; Dr. Vinay Kumar; Dr. Komal Malik; Dr. Tarandeep Singh Walia; and Deepender.
The application for the patent was published on May 1, under issue no. 18/2026.
According to the abstract released by the Intellectual Property India: "The current invention is associated with a federated Bayesian learning system of privacy-preserving distributed forecasting. The system allows the decentralized trainers of Bayesian models to so train locally and exchange posterior distributions rather than raw data. These updates are centralized at one aggregation server to create a global predictive model. The framework uses communication optimization and trust-based weighting as a means to enhance efficiency and strength. The invention offers scalable, privacy-aware, and uncertainty-sensitive predictive analytics which can be deployed in a distributed setting like healthcare, agriculture, and finance."
Disclaimer: Curated by HT Syndication.
