Fidelity-based supervised and unsupervised learning for binary classification of quantum states

Here, we develop two quantum-computational schemes for supervised and unsupervised classification tasks in a quantum world by employing the information-geometric tools of fidelity search algorithm. Presuming that pure states set given systems (or objects) belong to one known classes, objective here is decide which these classes each system belongs—without knowing its state. The binary algorithm based on having training sample whose class memberships are already known. algorithm, however, uses oracle knows membership computational basis. Both algorithms require ability evaluate between with unknown states, also general scheme.