Quantum semi-supervised kernel learning

Abstract

Quantum machine learning methods have the potential to facilitate learning using extremely large datasets. While the availability of data for training machine learning models is steadily increasing, oftentimes it is much easier to collect feature vectors to obtain the corresponding labels. One of the approaches for addressing this issue is to use semi-supervised learning, which leverages not only the labeled samples, but also unlabeled feature vectors. Here, we present a quantum machine learning algorithm for training semi-supervised kernel support vector machines. The algorithm uses recent advances in quantum sample-based Hamiltonian simulation to extend the existing quantum LS-SVM algorithm to handle the semisupervised term in the loss. Through a theoretical study of the algorithm's computational complexity, we show that it maintains the same speedup as the fully-supervised quantum LS-SVM.

Multi-Angle Fast Neural Tangent Kernel Classifier

“…It can provide a novel solution to process incomplete-view data and can also be adjusted to address large-scale complete-view learning problems efficiently. Saeedi [29] proposed an algorithm that uses recent advances in quantum sample-based Hamiltonian simulation to extend the existing quantum LS-SVM algorithm to handle the semi-supervised term in the loss. Guo et al [30] proposed a novel semi-supervised multiple empirical kernel learning (SSMEKL) which enables various practical kernel learning to achieve better classification performance with a small number of labeled samples and many unlabeled samples.…”

Multi-scale Attention Convolutional Neural Network for time series classification

“…In many practical machine learning tasks, C in and C out are often large, resulting in a huge number of quantum circuits and significantly increasing training time. Recently, Smaldone et al [50] proposed several quantum convolution methods that can process multi-channel data and require a number of qubits independent of the number of channels, while preserving inter-channel information. However, most of these methods require very deep quantum circuits, which is impractical for current NISQ devices.…”

“…Differing from previous works, we propose in this paper a novel QNN model based on quantum convolution. Quantum convolutional neural networks (QCNNs) were first introduced by Cong et al [39], which then triggered a series of research efforts on QCNNs [40][41][42][43][44][45][46][47][48][49][50][51][52]. However, most of these studies are concentrated on the application of QCNNs in the field of computer vision, with very limited exploration in the NLP domain.…”