Image Details

Choose export citation format:

Applying Deep Learning to Fast Radio Burst Classification

  • Authors: Liam Connor and Joeri van Leeuwen

2018 The Astronomical Journal 156 256.

  • Provider: AAS Journals

Caption: Figure 7.

Comparison of performance metrics between the four input data products using a subset of Apertif FRB candidates. As expected, the time/frequency array (dedispersed dynamic spectrum) has the most predictive power, and in this case is not significantly outperformed by the combined information. However, in special cases like multibeam peryton detections or low-DM events, the extra information is critical, even if they do not affect the overall statistics.

Other Images in This Article
Copyright and Terms & Conditions