Neural Network Approaches for Mobile Spectroscopic Gamma-Ray Source Detection

This paper investigates relatively simple (i.e., feedforward convolutional and MLP) neural networks for radiation detection. The main idea is to benchmark these networks against other established methods to assess the tradeoffs. In general, the performance was on par with the NMF-based method. Additionally, RNNs were used to add time dependence, which yielded even better results.

You can check the paper out here, and the PyTorch models used for the analysis in this paper can be found on my github profile.