Hybrid classical-quantum autoencoder for anomaly detection

We propose a hybrid classical-quantum autoencoder (HAE) model, which is synergy of classical (AE) and parametrized quantum circuit (PQC) that inserted into its bottleneck. The PQC augments the latent space by lifting it to whereby further data manipulations occur before performing measurement collapsing state original representation. From this resulting data, standard outlier detection method applied search for anomalous points within dataset. Using model applying both benchmarking datasets, specific use-case dataset, relates predictive maintenance gas power plants, we show addition bottleneck leads performance enhancement in terms precision, recall, F1 score. Furthermore, probe different Ansätze analyze features make them effective task.