Practical insights on the effect of different encodings, ansätze and measurements in quantum and hybrid convolutional neural networks
This work provides practical insights for designing quantum machine learning models, particularly for satellite image classification, but it is incremental as it focuses on benchmarking existing components rather than introducing new methods.
The study systematically evaluated the impact of data encoding, variational ansätze, and measurement strategies on quantum and hybrid convolutional neural networks for satellite image classification, finding that data encoding is the dominant factor in hybrid models with validation accuracy varying over 30%, while measurement protocols had the largest effect in purely quantum models with variations up to 30%.
This study investigates the design choices of parameterized quantum circuits (PQCs) within quantum and hybrid convolutional neural network (HQNN and QCNN) architectures, applied to the task of satellite image classification using the EuroSAT dataset. We systematically evaluate the performance implications of data encoding techniques, variational ansätze, and measurement in approx. 500 distinct model configurations. Our analysis reveals a clear hierarchy of influence on model performance. For hybrid architectures, which were benchmarked against their direct classical equivalents (e.g. the same architecture with the PQCs removed), the data encoding strategy is the dominant factor, with validation accuracy varying over 30% for distinct embeddings. In contrast, the selection of variational ansätze and measurement basis had a comparatively marginal effect, with validation accuracy variations remaining below 5%. For purely quantum models, restricted to amplitude encoding, performance was most dependent on the measurement protocol and the data-to-amplitude mapping. The measurement strategy varied the validation accuracy by up to 30% and the encoding mapping by around 8 percentage points.