During the last decade, considerable effort has been made to perform automatic classification of variable stars using machine learning techniques. Traditionally, light curves are represented as a vector of descriptors or features used as input for many algorithms. Some features are computationally expensive, cannot be updated quickly and hence for large datasets such as the LSST cannot be applied. Previous work has been done to develop alternative unsupervised feature extraction algorithms for light curves, but the cost of doing so still remains high. In this work, we propose an end-to-end algorithm that automatically learns the representation of light curves that allows an accurate automatic classification. We study a series of deep learning architectures based on Recurrent Neural Networks and test them in automated classification scenarios. Our method uses minimal data preprocessing, can be updated with a low computational cost for new observations and light curves, and can scale up to massive datasets. We transform each light curve into an input matrix representation whose elements are the differences in time and magnitude, and the outputs are classification probabilities. We test our method in three surveys: OGLE-III, Gaia and WISE. We obtain accuracies of about $95\%$ in the main classes and $75\%$ in the majority of subclasses. We compare our results with the Random Forest classifier and obtain competitive accuracies while being faster and scalable. The analysis shows that the computational complexity of our approach grows up linearly with the light curve size, while the traditional approach cost grows as $N\log{(N)}$.

The Galactic center (GC) is the densest region of the Milky Way. Variability surveys towards the GC potentially provide the largest number of variable stars per square degree within the Galaxy. However, high stellar density is also a drawback due to blending. Moreover, the GC is affected by extreme reddening, therefore near infrared observations are needed. We plan to detect new variable stars towards the GC, focusing on type II Cepheids (T2Cs) which have the advantage of being brighter than RR Lyrae stars. We perform parallel Lomb-Scargle and Generalized Lomb-Scargle periodogram analysis of the $K_s$-band time series of the VISTA variables in the Via Lactea survey, to detect periodicities. We employ statistical parameters to clean our sample. We take account of periods, light amplitudes, distances, and proper motions to provide a classification of the candidate variables. We detected 1,019 periodic variable stars, of which 164 are T2Cs, 210 are Miras and 3 are classical Cepheids. We also found the first anomalous Cepheid in this region. We compare their photometric properties with overlapping catalogs and discuss their properties on the color-magnitude and Bailey diagrams. We present the most extensive catalog of T2Cs in the GC region to date. Offsets in E($J-K_s$) and in the reddening law cause very large ($\sim$1-2 kpc) uncertainties on distances in this region. We provide a catalog which will be the starting point for future spectroscopic surveys in the innermost regions of the Galaxy.

The population of nearby dwarf galaxies in the Local Group constitutes a complete galactic environment, perfect suited for studying the connection between stellar populations and galaxy evolution. In this study, we are conducting an optical monitoring survey of the majority of dwarf galaxies in the Local Group, with the Isaac Newton Telescope (INT), to identify long period variable stars (LPVs). These stars are at the end points of their evolution and therefore their luminosity can be directly translated into their birth masses; this enables us to reconstruct the star formation history. By the end of the monitoring survey, we will have performed observations over ten epochs, spaced approximately three months apart, and identified long period, dust-producing AGB stars; five epochs of data have been obtained already. LPVs are also the main source of dust; in combination with Spitzer Space Telescope images at mid-IR wavelengths we will quantify the mass loss, and provide a detailed map of the mass feedback into the interstellar medium. We will also use the amplitudes in different optical passbands to determine the radius variations of the stars, and relate this to their mass loss.

We report 3 new R Coronae Borealis and 63 new DY Persei candidates in the Small Magellanic Cloud. Our analysis, based on data published by the OGLE team, consisted in a search for the characteristic drops in brightness that define these classes. All candidates had been previously classified as semi-regular or Mira variables. We briefly remark upon the possible existence of a "borderline" DY Per-like star and a "transitional" DY Per/RCB star. Follow-up observations are needed to conclusively establish the nature of our candidates.