Photon Number Classification

Comparison of different algorithms for the classification of transition edge sensor signals. With the development of a variety of techniques in the field of machine learning the goal is to quantify the advantages of modern classification techniques in the context of photon detection.

Files

Confidence

• Methods_Uniform.ipynb
• Methods_Geometric.ipynb
• Methods_Large.ipynb
• Methods_Noise.ipynb

Training

• Train_PtSNE.ipynb
• Train_PUMAP.ipynb

Figures

• Effect_Gaussians.ipynb
• Figures_General.ipynb
• Figures_Methods.ipynb
• Find_Distributions.ipynb

Experiments

The different algorithms are compared in a single notebook available in : Methods_Uniform.ipynb

The following methods are evaluated :

• Maximum Value
• Area
• Principal Component Analysis (PCA)
• Kernel Principal Component Analysis (K-PCA)
• t-Distributed Stochastic Neighbor Embedding (t-SNE)
• Uniform Manifold Approximation and Projection (UMAP)
• Non-Negative Matrix Factorization (NMF)
• Isomap
• Parametric t-SNE
• Parametric UMAP

Data Availability

All the data used in this research is available on the Zenodo open repository :

• Synthetic Uniform and Synthetic Geometric
• Synthetic Large and Noise

TODO

• Include Sphinx documentation

Acknowledgements

N.D.-C. and N.Q. acknowledge support from the Ministère de l'Économie et de l'Innovation du Québec, the Natural Sciences and Engineering Research Council Canada, Photonique Quantique Québec, and thank S. Montes-Valencia, J. Martinez-Cifuentes and A. Boon for valuable discussions. We also thank Z. Levine and S. Glancy for their careful feedback on our manuscript.