MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641104733 A) filed by Vardhaman College Of Engineering on August 31, 2026, for Neuro-Symbolic Ai For Real-Time Decision Making In Uncertain Environments.
Inventors include Ms B Swapna; Ms. P. Swetha; Mrs. Yadla Sunanda; Ms. Farhana Begum; Dr. Muni Sekhar Velpuru; and Dr. Vivek Kulkarni.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: Neuro-Symbolic AI for Real-Time Decision Making in uncertain Environments is the proposed invention. The proposed invention combines graph based relational learning with logical inference for processing heterogeneous data, modelling complex relationships among interconnected entities, and generating accurate, explainable and context-aware decisions. Firstly, the multi-source data are pre-processed and represented as a dynamic graph. The Graph Neural Network analyses the data to extract structural dependencies and contextual features. The learned representations are then fused with a symbolic reasoning engine that exploits domain knowledge, logical rules and explicit constraints to verify decisions and resolve ambiguities. An uncertainty management module estimates confidence levels for predictions and reasoning results in order to enable robust decision making even in the case of incomplete, noisy, or fast-changing information. Additionally, the framework is equipped with an explainability module that generates transparent, human-readable explanations for each decision. Furthermore, the framework incorporates a continual learning mechanism, which incrementally updates the graph learning model and the symbolic knowledge base using the newly obtained information.
Disclaimer: Curated by HT Syndication.
