Progress in quantum machine learning – HQS Quantum Simulations develops an innovative AutoSelection backend library for the AutoQML project.
As part of the AutoQML project on behalf of Fraunhofer IPA, HQS has developed a robust software library , the AutoSelectionBackend, which enables the automatic transpilation of quantum machine learning circuits. To simplify and optimize the development of QML solutions in an industrial context, the AutoQML project, led by Fraunhofer IAO, aims to integrate quantum components into the field of automated machine learning (AutoML). The library has been successfully integrated into the AutoQML framework and is now available as an open-source solution on the PlanQK platform.
Quantum computers offer enormous potential for accelerating machine learning approaches and developing new solutions. The AutoQML project focused on unlocking this potential and making quantum algorithms more accessible in practice. HQS's contribution on behalf of Fraunhofer IPA was to develop a robust software library. The challenge for HQS was to consider the quality of the quantum machines used while considering financial and time aspects.
The development of the AutoSelectionBackend library greatly benefits automated quantum machine learning. It makes it possible to automatically convert quantum machine learning circuits into a form suitable for quantum computers and to assign the qubits to the best quantum backends. This simplifies and optimizes the process of machine learning with quantum computers. The library considers various factors such as the quality of the quantum machines, the configuration of the backends, and the constantly changing noise models. This results in a more efficient use of quantum resources and enables users to achieve faster and more accurate results.
The AutoSelectionBackend library is a milestone that will benefit both newcomers and experts in quantum computing, simplifying access to this technology. For more information on our approach to the AutoQML project, please refer to the AutoQML case study.
More information:
https://platform.planqk.de/
https://www.autoqml.ai/