International Conference on Trends in Quantum Computing and Emerging Business Technologies (TQCEBT)
Quantum Information: Learning-Based Entanglement State Classification
Yashkumar Rakeshkumar Patel
Abstract
This paper proposes a novel method which employs machine learning to detect quantum entanglement in quantum states. A classical classification model is trained on a synthetic dataset, yielding a robust detector resilient to diverse noise profiles while maintaining high sample efficiency. K-local Pauli observables are used to gain insight into the classical shadow method. Numerical evaluations confirm that the model accurately identifies entangled multi-qubit states in noisy environments, facilitating the development of practical quantum information protocols.