Intellectual Property India Publishes Patent Application for An Ai And Ml System With Self-Evolving Neural Architecture For Automated Decision Making In Dynamic Environments

MUMBAI, India, Sept. 28 -- Intellectual Property India has published a patent application (202621100152 A) filed by Vishwakarma University, Pune on August 19, 2026, for An Ai And Ml System With Self-Evolving Neural Architecture For Automated Decision Making In Dynamic Environments.

Inventors include Mr. Tareek M. Pattewar; Mr. Vivek S. Thorat; Dr. Chetan Chauhan; Dr. Rahul R. Papalkar; Dr. Jayendra S. Jadhav; Mr. Harish S. Motekar; Dr. Mayur Deshmukh; Dr. Rajkumar Sadashiv Jagdale; Mr. Suhas Bhagwan Chavan; and Mr. Ajit Shankar Mali.

The application for the patent was published on September 25, 2026, under issue no. 39/2026.

Abstract: The present invention relates to an AI and ML system with a self-evolving neural architecture for automated decision making in dynamic environments. The system comprises a Data Acquisition Module, Data Preprocessing Module, Feature Processing Module, Environment Monitoring Module, Neural Architecture Module, Architecture Evolution Module, Adaptive Learning Module, Decision Generation Module, Performance Evaluation Module, Model Repository and Control Module. Environmental information is continuously acquired, validated and transformed into computational features for neural processing. The Environment Monitoring Module detects changes in input characteristics and operating conditions, while the Performance Evaluation Module determines corresponding model performance. Based on detected environmental changes and performance conditions, the Architecture Evolution Module selectively modifies structural characteristics of the neural architecture, and the Adaptive Learning Module updates associated learning parameters using current information. Candidate architectures are validated against performance and computational constraints before controlled deployment. The system further maintains architecture versions and enables rollback of previously validated configurations. The coordinated architecture provides adaptive neural computation and automated decision generation under changing environmental conditions.

Disclaimer: Curated by HT Syndication.