MUMBAI, India, Oct. 5 -- Intellectual Property India has published a patent application (202641115581 A) filed by Samreddy Pooja Reddy; Dr G. K. Srikanth; Vitteshkumar Dattatray Gaikwad; Dr. Battapotula Venkataratnam; Dr. Srikanth Kadainti; and Sagar Sathuluri on September 26, 2026, for Hybrid Ai Model Integrating Symbolic Reasoning And Deep Learning.
Inventors include Samreddy Pooja Reddy; Dr G. K. Srikanth; Vitteshkumar Dattatray Gaikwad; Dr. Battapotula Venkataratnam; Dr. Srikanth Kadainti; and Sagar Sathuluri.
The application for the patent was published on October 02, 2026, under issue no. 40/2026.
Abstract: A hybrid artificial intelligence system and method are disclosed that integrate a deep learning module with a symbolic reasoning engine through a bidirectional neuro- symbolic translation layer. The deep learning module computes a sub-symbolic representation and confidence measure from input data, which is translated into symbolic predicates for logical inference against a symbolic knowledge base comprising an ontology, rules, and facts, yielding an inferential result and a derivation trace. Symbolic constraints or query results may, in turn, be translated back into a differentiable signal that influences the deep learning module's processing or training, enabling bidirectional coupling between the two paradigms rather than a one-directional pipeline. A fusion and confidence arbitration module combines the sub- symbolic and symbolic outputs into a final decision, and an output generator produces the decision together with a human-readable explanation derived from the derivation trace. A feedback and continual learning module updates both the neural parameters and the symbolic knowledge base based on subsequently obtained outcome data, enabling the system's neural and symbolic components to co-evolve over its operational lifetime.
Disclaimer: Curated by HT Syndication.
