Quantum Self-Supervised Learning
This work addresses the scalability problem for machine learning practitioners by proposing a quantum approach, but it is incremental as it focuses on proof-of-principle experiments.
The paper tackles the computational bottleneck of self-supervised learning by exploring quantum neural networks, finding that small-scale quantum models show a numerical advantage in learning visual representations and achieve equal accuracy to classical models on downstream tasks using current noisy quantum hardware.
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.