MUMBAI, India, Aug. 17 -- Intellectual Property India has published a patent application (202641093916 A) filed by Vardhaman College Of Engineering on August 03, 2026, for Hybrid Quantum-Classical Machine Learning Model For Complex Optimization Problems.
Inventors include Dr. Lingam Sunitha; Ms. Dhanalaxmi Chinthala; Mr. Vinayak; Ms. Swetha Polisetty; Ms. Shaikh Sumaiya; and Mr. Satheesh Kumar S.
The application for the patent was published on August 14, 2026, under issue no. 33/2026.
Abstract: Hybrid Quantum-Classical Machine Learning Model for Complex Optimization Problems is the proposed invention. The present invention proposes a hybrid quantum- classical machine learning model to solve complex optimisation problems based on a GNN combined with a Variational Quantum Eigensolver (VQE). The optimisation variables and constraints are represented as a weighted graph, which allows the GNN to learn structural relationships, identify influential decision variables and reduce the search space before quantum processing. The reduced optimisation graph is mapped to parameterised quantum circuits, where the VQE efficiently searches candidate solutions in quantum superposition and a classical optimiser iteratively updates the circuit parameters until convergence. We present an adaptive orchestration engine that dynamically distributes computational tasks between classical and quantum resources according to problem complexity and hardware availability. An explainable optimisation module provides for transparent decision-making with confidence scores, node importance, and constraint satisfaction metrics. It results in faster convergence, better solution quality, lower computational overhead, efficient utilisation of quantum resources, and scalable deployment in logistics, finance, manufacturing, healthcare, telecommunications, and smart infrastructure optimisation.
Disclaimer: Curated by HT Syndication.
