Spectral Query
A spectral query is a task-definition input in quantum machine learning that specifies how a signal is probed. By modulating combinations of signal frequencies, it alters the hypothesis class of the circuit without introducing new sample information.
Timeline (last 90 days)
A study, by analyzing data re-uploading quantum circuits, revealed the mechanism by which the combination of signal frequencies is modulated through spectral queries defined by the encoding task.
Encoding spectral queries can change the hypothesis class of the circuit over the signal; the mechanism is modulating the way existing signal frequencies are combined, rather than introducing new sample-related information.
Experiments on small spectral tasks and radar-like tasks showed that, compared with circuits with matched parameters but mismatched or missing queries, circuits with aligned queries learn better from limited data.
All Facts (3)
A study, by analyzing data re-uploading quantum circuits, revealed the mechanism by which the combination of signal frequencies is modulated through spectral queries defined by the encoding task.
50%UnverifiedEncoding spectral queries can change the hypothesis class of the circuit over the signal; the mechanism is modulating the way existing signal frequencies are combined, rather than introducing new sample-related information.
50%UnverifiedExperiments on small spectral tasks and radar-like tasks showed that, compared with circuits with matched parameters but mismatched or missing queries, circuits with aligned queries learn better from limited data.
50%