Convolution filter embedded quantum gate autoencoder
Kodai Shiba,Katsuyoshi Sakamoto,Koichi Yamaguchi,Dinesh Bahadur Malla,Tomah Sogabe
DOI: https://doi.org/10.48550/arXiv.1906.01196
2019-06-04
Abstract:The autoencoder is one of machine learning algorithms used for feature extraction by dimension reduction of input data, denoising of images, and prior learning of neural networks. At the same time, autoencoders using quantum computers are also being developed. However, current quantum computers have a limited number of qubits, which makes it difficult to calculate big data. In this paper, as a solution to this problem, we propose a computation method that applies a convolution filter, which is one of the methods used in machine learning, to quantum computation. As a result of applying this method to a quantum autoencoder, we succeeded in denoising image data of several hundred qubits or more using only a few qubits under the autoencoding accuracy of 98%, and the effectiveness of this method was obtained. Meanwhile, we have verified the feature extraction function of the proposed autoencoder by dimensionality reduction. By projecting the MNIST data to two-dimension, we found the proposed method showed superior classification accuracy to the vanilla principle component analysis (PCA). We also verified the proposed method using IBM Q Melbourne and the actual machine failed to provide accurate results implying high error rate prevailing in the current NISQ quantum computer.
Quantum Physics,Emerging Technologies