Two-dimensional total absorption spectroscopy with conditional generative adversarial networks

Code

• github.com/alpha-davidson/sun_cgan
  OfficialIn papertf★ 0

Abstract

We explore the use of machine learning techniques to remove the response of large volume -ray detectors from experimental spectra. Segmented -ray total absorption spectrometers (TAS) allow for the simultaneous measurement of individual -ray energy (E_) and total excitation energy (E_x). Analysis of TAS detector data is complicated by the fact that the E_x and E_ quantities are correlated, and therefore, techniques that simply unfold using E_x and E_ response functions independently are not as accurate. In this work, we investigate the use of conditional generative adversarial networks (cGANs) to simultaneously unfold E_x and E_ data in TAS detectors. Specifically, we employ a Pix2Pix cGAN, a generative modeling technique based on recent advances in deep learning, to treat ~ matrix unfolding as an image-to-image translation problem. We present results for simulated and experimental matrices of single- and double- decay cascades. Our model demonstrates characterization capabilities within detector resolution limits for upwards of 93% of simulated test cases.

Tasks

Reproductions