Authors

Abstract

In astronomy, neural networks are often trained on simulation data with the prospect of being used on telescope observations.Unfortunately, training a model on simulation data and then applying it to instrument data leads to a substantial and potentiallyeven detrimental decrease in model accuracy on the new target dataset. Simulated and instrument data represent different datadomains, and for an algorithm to work in both, domain-invariant learning is necessary. Here we employ domain adaptationtechniques Maximum Mean Discrepancy (MMD) as an additional transfer loss and Domain Adversarial Neural Networks(DANNs) and demonstrate their viability to extract domain-invariant features within the astronomical context of classifyingmerging and non-merging galaxies. Additionally, we explore the use of Fisher loss and entropy minimization to enforce betterin-domain class discriminability. We show that the addition of each domain adaptation technique improves the performanceof a classifier when compared to conventional deep learning algorithms. We demonstrate this on two examples: between twoIllustris-1 simulated datasets of distant merging galaxies, and between Illustris-1 simulated data of nearby merging galaxies andobserved data from the Sloan Digital Sky Survey. The use of domain adaptation techniques in our experiments leads to an increaseof target domain classification accuracy of up to 20%. With further development, these techniques will allow astronomers tosuccessfully implement neural network models trained on simulation data to efficiently detect and study astrophysical objects incurrent and future large-scale astronomical surveys.