Quantum-Enhanced Spiking Neural Networks

This article explores the integration of Quantum Computing (QC) into classical Spiking Neural Networks (SNN) architectures to address limitations in encoding spiking signals into quantum circuits. Incorporating QC techniques to encode temporal dependencies within quantum circuits, the study aims to enhance the capabilities of SNN. It introduces a method lever-aging quantum systems' parallelism and high-dimensional state spaces to encode temporal information from classical spiking signals into a quantum framework relying on angle encoding and using two successive rotation gates. While the quantum circuit faced challenges in reconstructing complex temporal patterns, the proposed hybrid quantum-classical model, referred to as Quantum-enhanced Spiking Neural Networks (SQ-Net), significantly improved the efficiency and accuracy of processing temporal data using trigonometric functions. This advancement in temporal encoding within quantum circuits enables quantum systems to handle intricate temporal information effectively for sequential problems.