MUMBAI, India, July 30 -- Intellectual Property India has published a patent application (202641088571 A) filed by Madhankumar C; Arun R; S. Vijayakumari; Ms S Gokilavani; Dr. K. Murugeswari; Vahidhabanu Y; and Mrs. M. Indumathi on July 21, 2026, for Hybrid Neuro-Symbolic And Generative Artificial Intelligence Framework For Explainable Cognitive Computing With Self-Evolving Autonomous Decision Intelligence.
Inventors include Arun R; S. Vijayakumari; Ms S Gokilavani; Dr. K. Murugeswari; Vahidhabanu Y; and Mrs. M. Indumathi.
The application for the patent was published on July 24, 2026, under issue no. 30/2026.
Abstract: Hybrid Neuro-Symbolic and Generative Artificial Intelligence Framework for Explainable Cognitive Computing with Self-Evolving Autonomous Decision Intelligence Abstract The present invention discloses a Hybrid Neuro-Symbolic and Generative Artificial Intelligence Framework for Explainable Cognitive Computing with Self- Evolving Autonomous Decision Intelligence, integrating deep neural networks, symbolic reasoning, generative artificial intelligence, knowledge graphs, reinforcement learning, and explainable AI into a unified cognitive architecture. Conventional AI systems predominantly rely on data-driven neural models that often lack transparency, logical reasoning, and interpretability, while symbolic AI systems provide explainable reasoning but are limited in learning from complex, unstructured data. The proposed invention overcomes these limitations by combining neural perception, symbolic knowledge representation, and generative intelligence to enable transparent, context-aware, and autonomous decision-making with continuous self-improvement. The framework comprises a multimodal data acquisition module, neural intelligence engine, symbolic reasoning and knowledge graph module, generative AI engine, cognitive decision management module, explainability and validation engine, autonomous learning controller, adaptive knowledge repository, cloud edge computing infrastructure, and a self-evolving intelligence engine. The neural intelligence engine extracts semantic features from structured and unstructured data, including text, images, audio, video, IoT sensor streams, and enterprise databases, while the symbolic reasoning module applies logical inference, ontology-based reasoning, causal analysis, and rule validation to ensure consistency and reliability of AI-generated outputs. The generative AI engine synthesizes contextual knowledge, predicts future scenarios, generates recommendations, and supports autonomous planning. The explainability engine produces human-readable reasoning paths, confidence scores, evidence mapping, and decision traceability, thereby improving transparency, user trust, and regulatory compliance. The invention further incorporates reinforcement learning, federated learning, digital twin integration, cybersecurity mechanisms, and continuous feedback driven model evolution to enable adaptive intelligence across dynamic environments. The proposed framework is applicable to healthcare, smart manufacturing, autonomous robotics, cybersecurity, intelligent transportation, finance, education, scientific research, aerospace, digital twins, and sustainable smart infrastructure. By integrating neural learning, symbolic reasoning, generative artificial intelligence, and self-evolving autonomous decision intelligence within a single explainable cognitive computing platform, the invention significantly enhances reasoning accuracy, interpretability, adaptability, operational efficiency, and autonomous decision-making, thereby providing a scalable and trustworthy foundation for next-generation cognitive artificial intelligence systems.
Disclaimer: Curated by HT Syndication.
