Fast and efficient identification of anomalous galaxy spectra with neural density estimation

TL;DR

This paper demonstrates that an unsupervised probabilistic autoencoder can effectively identify unusual galaxy spectra in large datasets, outperforming traditional methods by capturing complex spectral distributions.

Contribution

It introduces the application of a Probabilistic Autoencoder for anomaly detection in galaxy spectra, enabling unsupervised identification of diverse and subtle spectral outliers.

Findings

Successfully detects various spectral outliers like E+A galaxies, LINERs, supernovae.

Lower probabilities assigned to spectra with unusual features.

Conditional modeling allows incorporation of additional information.

Abstract

Current large-scale astrophysical experiments produce unprecedented amounts of rich and diverse data. This creates a growing need for fast and flexible automated data inspection methods. Deep learning algorithms can capture and pick up subtle variations in rich data sets and are fast to apply once trained. Here, we study the applicability of an unsupervised and probabilistic deep learning framework, the Probabilistic Autoencoder (PAE), to the detection of peculiar objects in galaxy spectra from the SDSS survey. Different to supervised algorithms, this algorithm is not trained to detect a specific feature or type of anomaly, instead it learns the complex and diverse distribution of galaxy spectra from training data and identifies outliers with respect to the learned distribution. We find that the algorithm assigns consistently lower probabilities (higher anomaly score) to spectra that exhibit unusual features. For example, the majority of outliers among quiescent galaxies are E+A galaxies, whose spectra combine features from old and young stellar population. Other identified outliers include LINERs, supernovae and overlapping objects. Conditional modeling further allows us to incorporate additional information. Namely, we evaluate the probability of an object being anomalous given a certain spectral class, but other information such as metrics of data quality or estimated redshift could be incorporated as well. We make our code publicly available at https://github.com/VMBoehm/Spectra_PAE