The Inductive Bias of Quantum Kernels

Abstract

It has been hypothesized that quantum computers may lend themselves well to applications in machine learning. In the present work, we analyze function classes defined via quantum kernels. Quantum computers offer the possibility to efficiently compute inner products of exponentially large density operators that are classically hard to compute. However, having an exponentially large feature space renders the problem of generalization hard. Furthermore, being able to evaluate inner products in high dimensional spaces efficiently by itself does not guarantee a quantum advantage, as already classically tractable kernels can correspond to high-or infinite-dimensional reproducing kernel Hilbert spaces (RKHS).We analyze the spectral properties of quantum kernels and find that we can expect an advantage if their RKHS is low dimensional and contains functions that are hard to compute classically. If the target function is known to lie in this class, this implies a quantum advantage, as the quantum computer can encode this inductive bias, whereas there is no classically efficient way to constrain the function class in the same way. However, we show that finding suitable quantum kernels is not easy because the kernel evaluation might require exponentially many measurements.In conclusion, our message is a somewhat sobering one: we conjecture that quantum machine learning models can offer speed-ups only if we manage to encode knowledge about the problem at hand into quantum circuits, while encoding the same bias into a classical model would be hard. These situations may plausibly occur when learning on data generated by a quantum process, however, they appear to be harder to come by for classical datasets. * JMK and SB contributed equally and are ordered randomly.

Provable Advantage in Quantum Phase Learning via Quantum Kernel Alphatron

“…Tremendous classical kernel methods [45,46] have been proposed to learn the non-linear functions or decision boundaries. With the rapid development of quantum computers, there is a growing interest in exploring whether the quantum kernel method can surpass the classical kernel [35,36,39,[47][48][49][50][51][52][53][54][55][56][57][58][59][60][61]. Here we leverage the quantum kernel as our kernel function, which is defined as…”

Theory of overparametrization in quantum neural networks

“…Here, researchers have embedded ML into the framework of quantum mechanics, with the new, generalized theory being called Quantum Machine Learning (QML) [18][19][20]. With QML, the end goal is not formal generalization but rather to exploit entanglement and superposition to achieve a quantum advantage [21][22][23][24], that is, to solve the problem more efficiently than any classical algorithm run on a classical supercomputer.…”

Importance of Kernel Bandwidth in Quantum Machine Learning

“…This behavior is analogous to the curse of dimensionality in classical kernel methods 9 and not specific to the choice of fidelity as the similarity measure. One recently proposed approach to overcome this limitation is controlling the inductive bias of the quantum kernel methods by projecting the quantum state into a lower-dimensional subspace 7,10 . In general, however, additional information is required to appropriately choose the projection 10 .…”

“…One recently proposed approach to overcome this limitation is controlling the inductive bias of the quantum kernel methods by projecting the quantum state into a lower-dimensional subspace 7,10 . In general, however, additional information is required to appropriately choose the projection 10 .…”