Developing a Bubble Chamber Particle Discriminator Using Semi-Supervised Learning
Computational Physics
2018-11-29 v1 Instrumentation and Detectors
Abstract
The identification of non-signal events is a major hurdle to overcome for bubble chamber dark matter experiments such as PICO-60. The current practice of manually developing a discriminator function to eliminate background events is difficult when available calibration data is frequently impure and present only in small quantities. In this study, several different discriminator input/preprocessing formats and neural network architectures are applied to the task. First, they are optimized in a supervised learning context. Next, two novel semi-supervised learning algorithms are trained, and found to replicate the Acoustic Parameter (AP) discriminator previously used in PICO-60 with a mean of 97% accuracy.
Keywords
Cite
@article{arxiv.1811.11308,
title = {Developing a Bubble Chamber Particle Discriminator Using Semi-Supervised Learning},
author = {B. Matusch and C. Amole and M. Ardid and I. J. Arnquist and D. M. Asner and D. Baxter and E. Behnke and M. Bressler and B. Broerman and G. Cao and C. J. Chen and U. Chowdhury and K. Clark and J. I. Collar and P. S. Cooper and C. B. Coutu and C. Cowles and M. Crisler and G. Crowder and N. A. Cruz-Venegas and C. E. Dahl and M. Das and S. Fallows and J. Farine and I. Felis and R. Filgas and F. Girard and G. Giroux and J. Hall and C. Hardy and O. Harris and T. Hillier and E. W. Hoppe and C. M. Jackson and M. Jin and L. Klopfenstein and C. B. Krauss and M. Laurin and I. Lawson and A. Leblanc and I. Levine and C. Licciardi and W. H. Lippincott and B. Loer and F. Mamedov and P. Mitra and C. Moore and T. Nania and R. Neilson and A. J. Noble and P. Oedekerk and A. Ortega and M. -C. Piro and A. Plante and R. Podviyanuk and S. Priya and A. E. Robinson and S. Sahoo and O. Scallon and S. Seth and A. Sonnenschein and N. Starinski and I. Štekl and T. Sullivan and F. Tardif and E. Vázquez-Jáuregui and N. Walkowski and E. Weima and U. Wichoski and K. Wierman and Y. Yan and V. Zacek and J. Zhang},
journal= {arXiv preprint arXiv:1811.11308},
year = {2018}
}
Comments
27 pages, 10 figures