Deep Learning for Image Sequence Classification of Astronomical Events

We propose a new sequential classification model for astronomical objects based on a recurrent convolutional neural network (RCNN) which uses sequences of images as inputs. This approach avoids the computation of light curves or difference images. This is the first time that sequences of images are used directly for the classification of variable objects in astronomy. The second contribution of this work is the image simulation process. We generate synthetic image sequences which take into account the instrumental and observing conditions, obtaining a realistic, unevenly sampled, and variable noise set of movies for each astronomical object. The simulated data set is used to train our RCNN classifier. This approach allows us to generate data sets to train and test our RCNN model for different astronomical surveys and telescopes. Moreover, using a simulated data set is faster and more adaptable to different surveys and classification tasks. We aim to build a simulated data set whose distribution is close enough to the real data set, so that fine tuning could match the distributions. To test the RCNN classifier trained with the synthetic data set, we used real-world data from the High cadence Transient Survey (HiTS), obtaining an average recall of 85%, improved to 94% after performing fine tuning with 10 real samples per class. We compare the results of our RCNN model with those of a light curve random forest classifier. The proposed RCNN with fine tuning has a similar performance on the HiTS data set compared to the light curve random forest classifier, trained on an augmented training set with 10 real samples per class. The RCNN approach presents several advantages in an alert stream classification scenario, such as a reduction of the data pre-processing, faster online evaluation, and easier performance improvement using a few real data samples. The results obtained encourage us to use the proposed method for astronomical alert broker systems that will process alert streams generated by new telescopes such as the Large Synoptic Survey Telescope.

[1]  Eduardo Serrano,et al.  LSST: From Science Drivers to Reference Design and Anticipated Data Products , 2008, The Astrophysical Journal.

[2]  Pavlos Protopapas,et al.  The High Cadence Transit Survey (HiTS): Compilation and Characterization of Light-curve Catalogs , 2018, The Astronomical Journal.

[3]  N. A. Walton,et al.  The delay of shock breakout due to circumstellar material evident in most type II supernovae , 2018, Nature Astronomy.

[4]  Pablo A. Estévez,et al.  Clustering of Astronomical Transient Candidates Using Deep Variational Embedding , 2018, 2018 International Joint Conference on Neural Networks (IJCNN).

[5]  Pablo A. Estévez,et al.  Enhanced Rotational Invariant Convolutional Neural Network for Supernovae Detection , 2018, 2018 International Joint Conference on Neural Networks (IJCNN).

[6]  F. Förster,et al.  Discovery of Distant RR Lyrae Stars in the Milky Way Using DECam , 2018, 1802.01581.

[7]  C. Scheidegger,et al.  Machine-learning-based Brokers for Real-time Classification of the LSST Alert Stream , 2018, 1801.07323.

[8]  Pavlos Protopapas,et al.  Uncertain Classification of Variable Stars: Handling Observational GAPS and Noise , 2017, 1801.09732.

[9]  Christopher J. Shallue,et al.  Identifying Exoplanets with Deep Learning: A Five-planet Resonant Chain around Kepler-80 and an Eighth Planet around Kepler-90 , 2017, 1712.05044.

[10]  Brett Naul,et al.  A recurrent neural network for classification of unevenly sampled variable stars , 2017, Nature Astronomy.

[11]  Daniel George,et al.  Deep Learning for Real-time Gravitational Wave Detection and Parameter Estimation with Advanced LIGO Data , 2017, ArXiv.

[12]  Ashish Mahabal,et al.  Effective Image Differencing with ConvNets for Real-time Transient Hunting , 2017, ArXiv.

[13]  Pavlos Protopapas,et al.  Robust Period Estimation Using Mutual Information for Multiband Light Curves in the Synoptic Survey Era , 2017, ArXiv.

[14]  Patrick van der Smagt,et al.  Two-stream RNN/CNN for action recognition in 3D videos , 2017, 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[15]  Xiaolin Hu,et al.  Recurrent convolutional neural network for speech processing , 2017, 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[16]  Jianhua Z. Huang,et al.  The M33 Synoptic Stellar Survey. II. Mira Variables , 2017, 1703.01000.

[17]  Aniruddha R. Thakar,et al.  Sloan Digital Sky Survey IV: Mapping the Milky Way, Nearby Galaxies, and the Distant Universe , 2017, 1703.00052.

[18]  Sergey Ioffe,et al.  Batch Renormalization: Towards Reducing Minibatch Dependence in Batch-Normalized Models , 2017, NIPS.

[19]  Pablo A. Estévez,et al.  Deep-HiTS: Rotation Invariant Convolutional Neural Network for Transient Detection , 2017, ArXiv.

[20]  P. Mazzali,et al.  Light-curve and spectral properties of ultrastripped core-collapse supernovae leading to binary neutron stars , 2016, 1612.02882.

[21]  Meng Zhang,et al.  Neural Network Methods for Natural Language Processing , 2017, Computational Linguistics.

[22]  W. M. Wood-Vasey,et al.  The Pan-STARRS1 Surveys , 2016, 1612.05560.

[23]  F. Förster,et al.  THE HIGH CADENCE TRANSIENT SURVEY (HITS). I. SURVEY DESIGN AND SUPERNOVA SHOCK BREAKOUT CONSTRAINTS , 2016, 1609.03567.

[24]  Pablo A. Estévez,et al.  Supernovae detection by using convolutional neural networks , 2016, 2016 International Joint Conference on Neural Networks (IJCNN).

[25]  Tom Charnock,et al.  Deep Recurrent Neural Networks for Supernovae Classification , 2016, ArXiv.

[26]  Tara N. Sainath,et al.  Convolutional, Long Short-Term Memory, fully connected Deep Neural Networks , 2015, 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[27]  Pavlos Protopapas,et al.  FATS: Feature Analysis for Time Series , 2015, 1506.00010.

[28]  Zachary Chase Lipton.  A Critical Review of Recurrent Neural Networks for Sequence Learning , 2015, ArXiv.

[29]  Trevor Darrell,et al.  Long-term recurrent convolutional networks for visual recognition and description , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[30]  Gerald Penn,et al.  Convolutional Neural Networks for Speech Recognition , 2014, IEEE/ACM Transactions on Audio, Speech, and Language Processing.

[31]  Pavlos Protopapas,et al.  Computational Intelligence Challenges and Applications on Large-Scale Astronomical Time Series Databases , 2014, IEEE Computational Intelligence Magazine.

[32]  Christopher Bebek,et al.  The Zwicky Transient Facility: Observing System , 2014, Astronomical Telescopes and Instrumentation.

[33]  Ivan Laptev,et al.  Learning and Transferring Mid-level Image Representations Using Convolutional Neural Networks , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[34]  M. Feast,et al.  Cepheid variables in the flared outer disk of our galaxy , 2014, Nature.

[35]  Pavlos Protopapas,et al.  The EPOCH Project - I. Periodic variable stars in the EROS-2 LMC database , 2014, 1403.6131.

[36]  Pavlos Protopapas,et al.  AUTOMATIC CLASSIFICATION OF VARIABLE STARS IN CATALOGS WITH MISSING DATA , 2013, ArXiv.

[37]  Geoffrey E. Hinton,et al.  Speech recognition with deep recurrent neural networks , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[38]  M. Wainwright,et al.  Using machine learning for discovery in synoptic survey imaging data , 2012, 1209.3775.

[39]  E. O. Ofek,et al.  Automating Discovery and Classification of Transients and Variable Stars in the Synoptic Survey Era , 2011, 1106.5491.

[40]  H. Thomas Diehl,et al.  The Dark Energy Survey Camera (DECam) , 2012 .

[41]  Linhua Jiang,et al.  LIGHT CURVE TEMPLATES AND GALACTIC DISTRIBUTION OF RR LYRAE STARS FROM SLOAN DIGITAL SKY SURVEY STRIPE 82 , 2009, 0910.4611.

[42]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[43]  Geoffrey E. Hinton,et al.  Visualizing Data using t-SNE , 2008 .

[44]  M. Sullivan,et al.  K-Corrections and Spectral Templates of Type Ia Supernovae , 2007, astro-ph/0703529.

[45]  J. Beaulieu,et al.  Deep Canada–France–Hawaii Telescope photometric survey of the entire M33 galaxy – I. Catalogue of 36 000 variable point sources★ , 2006 .

[46]  Christopher K. I. Williams,et al.  Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning) , 2005 .

[47]  V. Belokurov,et al.  Light-curve classification in massive variability surveys - II. Transients towards the Large Magellanic Cloud , 2004, astro-ph/0404232.

[48]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[49]  Kunihiko Fukushima,et al.  Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position , 1980, Biological Cybernetics.

[50]  Yann Le Du,et al.  Lightcurve Classification in Massive Variability Surveys , 2003 .

[51]  N. Wyn Evans,et al.  Light-curve classification in massive variability surveys — I. Microlensing , 2002, astro-ph/0211121.

[52]  A. G. Alexei,et al.  OBSERVATIONAL EVIDENCE FROM SUPERNOVAE FOR AN ACCELERATING UNIVERSE AND A COSMOLOGICAL CONSTANT , 1998 .

[53]  M. Phillips,et al.  The High-Z Supernova Search: Measuring Cosmic Deceleration and Global Curvature of the Universe Using Type Ia Supernovae , 1998, astro-ph/9805200.

[54]  T. Naylor.  An optimal extraction algorithm for imaging photometry , 1998 .

[55]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[56]  N. Lomb.  Least-squares frequency analysis of unequally spaced data , 1976 .