Hopfield Neural Network implementation with 3-SAT and Grover’s Algorithm

The objective of this work is to explore quantum computing methods to increase the computational speed of Hopfield Neural Network (HNN) in detection of a radioactive anomaly. The approach is based on the Grover’s search algorithm in conjunction with a 3-SAT problem formalism. The Grover’s algorithm is implemented on a quantum computing simulator using Qiskit software. Performance of HNN algorithm is benchmarked using search data from an environmental screening campaign, where the anomaly is a subset of measurements containing a ¹³⁷Cs source. Results indicate that using Grover’s algorithm on a quantum simulator reduces runtime of HNN by two orders of magnitude relative to classical HNN. This work has applications to environmental screening of gamma radiation, which involves rapid detection of weak nuisance and anomaly signal in the presence of strong and highly varying background.