Quantum classifiers for domain adaptation

Transfer learning (TL), a crucial subfield of machine learning, aims to accomplish task in the target domain with acquired knowledge source domain. Specifically, effective adaptation (DA) facilitates delivery TL where all data samples two domains are distributed same feature space. In this paper, quantum implementations DA classifier presented speedup compared classical classifier. One implementation, basic linear algebra subroutines-based classifier, can predict labels logarithmic resources number and dimension given data. The other implementation efficiently accomplishes through variational hybrid quantum-classical procedure.