MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541122806 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on Dec. 5, 2025, for 'climate pattern prediction system using meta-learning algorithms.'
Inventor(s) include Dr. T. Joshva Devadas; Dr. Naveenkumar Jayakumar; Chopade Yash Vishnu; and Raj Gupta.
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 present disclosure provides a climate pattern prediction system comprising a data collection module configured to acquire climate data from multiple sources including satellite data, meteorological station data, and real-time sensor networks, wherein the climate data includes temperature, precipitation, atmospheric pressure, humidity, and wind speed measurements across different geographical regions and temporal periods; a meta-learning training module configured to implement Model-Agnostic Meta-Learning algorithms that train base models on multiple climate prediction tasks from different regions and periods, wherein the algorithms learn initialization parameters enabling rapid adaptation to new tasks with minimal training data; and a prediction engine configured to generate climate forecasts by adapting the trained meta-learning model to new conditions using the learned parameters. The system enables rapid adaptation to new climate prediction scenarios through few-shot learning while maintaining accuracy across diverse geographical regions and temporal scales."
Disclaimer: Curated by HT Syndication.
