Self-supervised learning (SSL) has emerged as a novel method for learning representations from inputs without data annotations. A key challenge in SSL is avoiding the collapse of representations, leading to poor predictive models. This research proposes an SSL framework that leverages quantum classifiers to address this issue. We achieve well-stratified, nontrivial representations by embedding data into the Hilbert space via quantum classifiers. The algorithm works by minimizing the Kullback-Leibler (KL) divergence to maximize the agreement of positive pairs and minimize the agreement of negative pairs. We demonstrate the proof-of-concept in image classification tasks using MNIST, KMNIST, and FashionMNIST datasets. Classifiers trained with our framework achieve higher predictive performance than classical classifiers, with a maximum accuracy of 65% on MNIST and more than 70% on KMNIST and FashionMNIST using unseen samples without fine-tuning. The code implementation is available at https://github.com/namhai03/QSSL
Self-supervised learning (SSL) has emerged as a novel method for learning representations from inputs without data annotations. A key challenge in SSL is avoiding the collapse of representations, leading to poor predictive models. This research proposes an SSL framework that leverages quantum classifiers to address this issue. We achieve well-stratified, nontrivial representations by embedding data into the Hilbert space via quantum classifiers. The algorithm works by minimizing the Kullback-Leibler (KL) divergence to maximize the agreement of positive pairs and minimize the agreement of negative pairs. We demonstrate the proof-of-concept in image classification tasks using MNIST, KMNIST, and FashionMNIST datasets. Classifiers trained with our framework achieve higher predictive performance than classical classifiers, with a maximum accuracy of 65% on MNIST and more than 70% on KMNIST and FashionMNIST using unseen samples without fine-tuning. The code implementation is available at https://github.com/namhai03/QSSL
The resurgence of self-supervised learning, whereby a deep learning model generates its own supervisory signal from the data, promises a scalable way to tackle the dramatically increasing size of real-world data sets without human annotation. However, the staggering computational complexity of these methods is such that for state-of-the-art performance, classical hardware requirements represent a significant bottleneck to further progress. Here we take the first steps to understanding whether quantum neural networks could meet the demand for more powerful architectures and test its effectiveness in proof-of-principle hybrid experiments. Interestingly, we observe a numerical advantage for the learning of visual representations using small-scale quantum neural networks over equivalently structured classical networks, even when the quantum circuits are sampled with only 100 shots. Furthermore, we apply our best quantum model to classify unseen images on the ibmq\_paris quantum computer and find that current noisy devices can already achieve equal accuracy to the equivalent classical model on downstream tasks.
Abstract The resurgence of self-supervised learning, whereby a deep learning model generates its own supervisory signal from the data, promises a scalable way to tackle the dramatically increasing size of real-world data sets without human annotation. However, the staggering computational complexity of these methods is such that for state-of-the-art performance, classical hardware requirements represent a significant bottleneck to further progress. Here we take the first steps to understanding whether quantum neural networks (QNNs) could meet the demand for more powerful architectures and test its effectiveness in proof-of-principle hybrid experiments. Interestingly, we observe a numerical advantage for the learning of visual representations using small-scale QNN over equivalently structured classical networks, even when the quantum circuits are sampled with only 100 shots. Furthermore, we apply our best quantum model to classify unseen images on the ibmq_paris quantum computer and find that current noisy devices can already achieve equal accuracy to the equivalent classical model on downstream tasks.
ABSTRACT As an unsupervised feature representation paradigm, Self‐Supervised Learning (SSL) uses the intrinsic structure of data to extract meaningful features without relying on manual annotation. Despite the success of SSL, there are still problems, such as limited model capacity or insufficient representation ability. Quantum SSL has become a promising alternative because it can exploit quantum states to enhance expression ability and learning efficiency. This letter proposes a Quantum SSL with entanglement augmentation method (QSEA). Different from existing Quantum SSLs, QSEA introduces an entanglement‐based sample generation scheme and a fidelity‐driven quantum loss function. Specifically, QSEA constructs augmented samples by entangling an auxiliary qubit with the raw state and applying parameterized unitary transformations. The loss function is defined using quantum fidelity, quantifying similarity between quantum representations and effectively capturing sample relations. Experimental results show that QSEA outperforms existing quantum self‐supervised methods on multiple benchmarks and shows stronger stability in decorrelation noise environments. This framework lays the theoretical and practical foundation for quantum learning systems and advances the development of quantum machine learning in SSL.
Quantum machine learning (QML) has attracted growing interest with the rapid parallel advances in large-scale classical machine learning and quantum technologies. Similar to classical machine learning, QML models also face challenges arising from the scarcity of labeled data, particularly as their scale and complexity increase. Here, we introduce self-supervised pretraining of quantum representations that reduces reliance on labeled data by learning invariances from unlabeled examples. We implement this paradigm on a programmable trapped-ion quantum computer, encoding images as quantum states. In situ contrastive pretraining on hardware yields a representation that, when fine-tuned, classifies image families with higher mean test accuracy and lower run-to-run variability than models trained from random initialization. Performance improvement is especially significant in regimes with limited labeled training data. We show that the learned invariances generalize beyond the pretraining image samples. Unlike prior work, our pipeline derives similarity from measured quantum overlaps and executes all training and classification stages on hardware. These results establish a label-efficient route to quantum representation learning, with direct relevance to quantum-native datasets and a clear path to larger classical inputs.
In the evolving landscape of data privacy regulations, the challenge of providing extensive data for robust deep classification models arises. The accuracy of these models relies on the amount of training data, due to the multitude of parameters that require tuning. Unfortunately, obtaining such ample data proves challenging, particularly in domains like medical applications, where there is a pressing need for robust models for early disease detection but a shortage of labeled data. Nevertheless, the classical supervised contrastive learning models, have shown the potential to address this challenge up to a certain limit, by utilizing deep encoder models. However, recent advancements in quantum machine learning enable the extraction of meaningful representations from extremely limited and simple data. Thus, replacing classical counterparts in classical or hybrid quantum-classical supervised contrastive models enhances feature learning capability with minimal data. Therefore, this work proposes the Q-SupCon model, a fully quantum-powered supervised contrastive learning model comprising a quantum data augmentation circuit, quantum encoder, quantum projection head, and quantum variational classifier, enabling efficient image classification with minimal labeled data. Furthermore, the novel model attains 80%, 60%, and 80% test accuracy on MNIST, KMNIST, and FMNIST datasets, marking a significant advancement in addressing the data scarcity challenge.
Recently, Lloyd et al. [arXiv:2001.03622] introduced \textit{quantum metric learning} as a powerful framework for quantum supervised learning. Such strategy focuses on the training of the quantum embedding, with a goal of `well-separation' of the embedded data in the Hilbert space H. Motivated by such an idea, we present a metric-based framework for supervised learning with trainable quantum circuits. We introduce both explicit and implicit approaches, based on the \textit{metric} defined on the Hilbert space H, to classify data. In particular, with the implicit approach, the number of separated classes (or their labels) in supervised learning problems can be arbitrarily high (with respect to the number of given qubits, or more precisely, the number of measurement), which is an improvement over current quantum supervised learning models, and this approach requires relatively small number of training points. With the explicit approach, we establish an equivalence between our metric-based framework and other quantum supervised learning models. Such equivalence provides a unification of supervised learning frameworks on quantum computer. We also demonstrate our framework by performing both noise-free and noisy numerical simulations and by running the classification on real IBM Q devices.