QML for FRB classification

[NicoRenaud]

Supervised https://arxiv.org/abs/2210.02471 and unsupervised https://arxiv.org/abs/2210.02463 methods have been proposed to classify FRB. Can we use QML for that ?

We can reuse the

• quantum SVM : https://github.com/NLESC-quantum/quantum_comp/blob/main/pulsar_search/ml/kernel/kernel_pulsar_qiskit.ipynb
• 1Qbit classifier : https://github.com/NLESC-quantum/quantum_comp/blob/main/pulsar_search/ml/reupload/reupload_pennylane.ipynb

We had excellent result with the 1Qbit classifier for pulsar detection so that should be tried first. TNO is also using it in one of their project and have a robust code they want to open source and put on qalcore. We can ask them what the status is so that we don't duplicate work.

[NicoRenaud]

A paper about DL for FRBs done at Astron as well : https://arxiv.org/pdf/1803.03084.pdf They have synthetic data set that we are trying to get now