MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202611064634 A) filed by Geeta University on May 22, 2026, for “neocortex-Inspired Self-Configuring Ai (nsca) For Lifelong Sequential Task Learning”.
Inventors include Mr. Anurag Vashist; Dr. Kapil Saini; Dr. Poonam; Dr. Rekha Narang; and Mr. Pankaj Bajaj.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: NEOCORTEX-INSPIRED SELF-CONFIGURING AI (NSCA) FOR LIFELONG SEQUENTIAL TASK LEARNING Abstract of the Invention The invention relates to a Neocortex-Inspired Self-Configuring AI (NSCA) system for lifelong sequential task learning without catastrophic forgetting. The system comprises cortical column modules (CCMs), a meta-controller, a dynamic routing layer, a memory bank, and a skill composition engine. The meta-controller analyzes incoming tasks and selectively activates or reuses CCMs, while the routing layer forms sparse task-specific pathways. The memory bank stores compressed task representations for retrieval and reuse, enabling continual learning. The skill composition engine integrates modular outputs to solve complex or novel tasks. A feedback and self-consolidation mechanism refines active pathways while preserving prior knowledge. The proposed architecture enables scalable, reusable, and efficient learning across sequential tasks, supporting cross-domain adaptability and eliminating the need for retraining or catastrophic forgetting.
Disclaimer: Curated by HT Syndication.
