MUMBAI, India, Oct. 26 -- Intellectual Property India has published a patent application (202621086715 A) filed by Karunasagar Sridharan on July 15, 2026, for A System And Method For Universal Online Time-Series Prediction Using A Recursive Adaptive Process.
Inventor includes Karunasagar Sridharan.
The application for the patent was published on October 23, 2026, under issue no. 43/2026.
Abstract: A system and method for universal online time-series prediction using a recursive adaptive process. The system comprises an input module to receive a sequential time- series data stream. A forecaster module processes each data point row-by-row in a single online pass without storing historical data. A Hyfa module implements a recursive computational structure colony of single computational structures, each maintaining a finite set of persistent memory slots and executing a predetermined sequence of normalized weighted blending operations in one forward pass to generate both a predicted value and a confidence measure without training epochs or back- propagation. A simulation controller module executes gated self-optimization simulations when confidence falls below a threshold, injecting superior structural variants at a reduced learning rate. An output module generates a sequence of predicted future values in real time, enabling deployment on microcontrollers without a training phase or retraining overhead.
Disclaimer: Curated by HT Syndication.
