Quantum active learning
Phys. Rev. A 112, 012409 – Published 10 July, 2025
DOI: https://doi.org/10.1103/c8q9-6vy7
Abstract
Quantum machine learning (QML), as an extension of classical machine learning that harnesses quantum mechanics, facilitates efficient learning from data encoded in quantum states. Training a quantum neural network (QNN) typically demands a substantial labeled training set for supervised learning. Human annotators, often experts, provide labels for samples through additional experiments, adding to the training cost. To mitigate this expense, there is a quest for methods that maintain model performance over fully labeled data sets while requiring fewer labeled samples in practice, thereby extending few-shot learning to the quantum realm. Quantum active learning (QAL) estimates the uncertainty of quantum data to select the most informative samples from a pool for labeling. Consequently, a QML model is supposed to accumulate maximal knowledge as the training set comprises labeled samples selected via sampling strategies. Notably, the QML models trained within the QAL framework are not restricted to specific types, allowing performance enhancement from the model architecture's perspective towards few-shot learning. Recognizing symmetry as a fundamental concept in physics ubiquitous across various domains, we leverage the symmetry inherent in quantum states induced by the embedding of classical data for model design. We employ an equivariant QNN capable of generalizing from fewer data with geometric priors. We benchmark the performance of QAL on two classification problems, observing both positive and negative results. QAL effectively trains the model, achieving performance comparable to that on fully labeled data sets by labeling less than 7% of the samples in the pool with unbiased sampling behavior. Furthermore, we elucidate the negative result of QAL being overtaken by random sampling baseline through miscellaneous numerical experiments. Our work lays the groundwork for real-world applications of QML by addressing realistic experimental costs.
