Table 6 Comparison to other algorithmic, ANN, and SNN baselines on the LOFAR dataset

From: Spiking neural networks for radio frequency interference detection in radio astronomy

Work

Model

Method

AUROC

AUPRC

F1

15

AOFlagger

Algorithm

0.788

0.572

0.570

18

Auto-Encoder

ANN

  

0.742

16

Auto-Encoder

ANN

0.989

0.748

0.660

25

Auto-Encoder

ANN2SNN

0.609

0.321

0.408

This work

BPTT

SNN

0.346

0.604

0.474

  1. Abbreviations: DN divisive normalisation, SNN spiking neural network, ANN artificial neural network, BPTT backpropagation through time, AUROC area under the receiver operator curve, and AUPRC area under the precision recall curve. While performance still lags the state of the art, these results encourage further development considering the compact size of our BPTT trained SNNs.
Back to article page