AlbaNova A3:1003 - Styrelserum/KOMKO (AlbaNova Main Building)
AlbaNova A3:1003 - Styrelserum/KOMKO
AlbaNova Main Building
18
Description
Quantum kernel machine learning with continuous variables.
The popular qubit framework has dominated recent work on quantum kernels, with results characterising expressability, learnability and generalisation. As yet, there is no comparative framework to understand these concepts for continuous variable (CV) quantum computing platforms. In this paper we represent CV quantum kernels as holomorphic functions and use this representation to provide several theoretical insights. The approach permits a general closed form solution for all CV quantum kernels and shows every such kernel can be expressed as the product of Gaussian and polynomial terms. Furthermore, it enables quantification of a quantum-classical separation for all such kernels via a notion of "stellar rank", and provides intuition for how bandwidth hyper-parameter tuning results in trades-off between learnability and efficient classical simulability.
