On Optimizing Hyperparameters for Quantum Neural Networks
Sabrina Herbst, Vincenzo De Maio, Ivona Brandic
TL;DR
This work tackles the challenge of hyperparameter tuning in quantum neural networks (QNNs) by conducting an extensive empirical study across four classical datasets using IBM Qiskit. It systematically varies data encoding, optimizers, ansatz structures, entanglement patterns, and preprocessing, under both noiseless and noisy simulations. The key finding is that optimizer choice and initialization dominate trainability, with beta-distributed initialization often improving performance, especially for entangled feature maps, while entanglement strategy shows limited consistent impact. The study provides concrete guidelines and publicly available data to help researchers and practitioners design more trainable QNNs for near-term hardware, signaling practical pathways for accelerating quantum-assisted ML in the Post-MMoore era.
Abstract
The increasing capabilities of Machine Learning (ML) models go hand in hand with an immense amount of data and computational power required for training. Therefore, training is usually outsourced into HPC facilities, where we have started to experience limits in scaling conventional HPC hardware, as theorized by Moore's law. Despite heavy parallelization and optimization efforts, current state-of-the-art ML models require weeks for training, which is associated with an enormous $CO_2$ footprint. Quantum Computing, and specifically Quantum Machine Learning (QML), can offer significant theoretical speed-ups and enhanced expressive power. However, training QML models requires tuning various hyperparameters, which is a nontrivial task and suboptimal choices can highly affect the trainability and performance of the models. In this study, we identify the most impactful hyperparameters and collect data about the performance of QML models. We compare different configurations and provide researchers with performance data and concrete suggestions for hyperparameter selection.