Identifying strong lenses with unsupervised machine learning using convolutional autoencoder

  title={Identifying strong lenses with unsupervised machine learning using convolutional autoencoder},
  author={Ting-Yun Cheng and Nan Li and Christopher J. Conselice and Alfonso Arag'on-Salamanca and Simon Dye and R. Benton Metcalf},
  journal={Monthly Notices of the Royal Astronomical Society},
In this paper, we develop a new unsupervised machine learning technique comprised of a feature extractor, a convolutional autoencoder, and a clustering algorithm consisting of a Bayesian Gaussian mixture model. We apply this technique to visual band space-based simulated imaging data from the Euclid Space Telescope using data from the strong gravitational lenses finding challenge. Our technique promisingly captures a variety of lensing features such as Einstein rings with different radii… 

Beyond the hubble sequence – exploring galaxy morphology with unsupervised machine learning

The main result in this work is not how well the unsupervised method matches visual classifications and physical properties, but that the method provides an independent classification that may be more physically meaningful than any visually based ones.

Machine learning in astronomy

This paper describes the use of machine learning and deep learning in astronomy through the examples of star-galaxy classification and the classification of low-mass X-ray binaries into binaries, which host a neutron star and those whichHost a black hole.

Applications of Machine Learning to Predicting Core-collapse Supernova Explosion Outcomes

A novel look at identifying the explosion outcome of core-collapse supernovae using a machine-learning approach is presented and it is found that the density profiles alone contain meaningful information regarding their explodability.

An Unsupervised Hunt for Gravitational Lenses

A lens detection method that combines simu-lation, data augmentation, semi-supervised learning, and GANs to improve this performance by an order of magnitude and allow researchers to go into a survey mostly “blind” and still classify strong gravitational lenses with high precision and recall.

Automatic Morphological Classification of Galaxies: Convolutional Autoencoder and Bagging-based Multiclustering Model

In order to obtain morphological information of unlabeled galaxies, we present an unsupervised machine-learning (UML) method for morphological classification of galaxies, which can be summarized as

Mining for Strong Gravitational Lenses with Self-supervised Learning

This work presents 1192 new strong lens candidates that are identified through a brief visual identification campaign, and releases an interactive web-based similarity search tool and the top network predictions to facilitate crowd-sourcing rapid discovery of additional strong gravitational lenses and other rare objects.

AI-driven spatio-temporal engine for finding gravitationally lensed supernovae

We present a spatio-temporal AI framework that concurrently exploits both the spatial and time-variable features of gravitationally lensed supernovae in optical images to ultimately aid in the

Lenses In VoicE (LIVE): searching for strong gravitational lenses in the VOICE@VST survey using convolutional neural networks

We present a sample of 16 likely strong gravitational lenses identified in the VST Optical Imaging of the CDFS and ES1 fields (VOICE survey) using Convolutional Neural Networks (CNNs). We train two

Strong lens modelling: comparing and combining Bayesian neural networks and parametric profile fitting

The vast quantity of strong galaxy-galaxy gravitational lenses expected by future large-scale surveys necessitates the development of automated methods to efficiently model their mass profiles. For



The use of the area under the ROC curve in the evaluation of machine learning algorithms


Ongoing and future imaging surveys represent significant improvements in depth, area, and seeing compared to current data sets. These improvements offer the opportunity to discover up to three orders

Pattern Recognition and Machine Learning (Informa

  • 2006

2017, in ICONIP

  • Hamana T., et al.,
  • 2004

Modelos de mezclas Bernoulli con regresión logística: una aplicación en la valoración de carteras de crédito

Este trabajo final de maestria, modalidad de profundizacion, consiste en la elaboracion de un problema de modelacion estadistica aplicada al sector crediticio. El objetivo es aplicar un modelo de

Predicción y selección de variables con bosques aleatorios en presencia de variables correlacionadas

El presente trabajo aborda el problema de seleccion de variables empleando el metodo de bosques aleatorios cuando el modelo subyacente para la variable respuesta es de tipo lineal. Para ello se

Galaxy morphological classification in deep-wide surveys via unsupervised machine learning

This study implements an algorithm that performs clustering of graph representations, in order to group image patches with similar visual properties and objects constructed from those patches, like galaxies, to autonomously reduce the galaxy population to a small number of ‘morphological clusters’.

Optimizing automatic morphological classification of galaxies with machine learning and deep learning using Dark Energy Survey imaging

A comparison between several common machine learning methods for galaxy classification by using Dark Energy Survey data combined with visual classifications from the Galaxy Zoo 1 project, showing that CNN is the most successful method of these ten methods in this study.

Self-Organizing Maps, Second Edition

  • T. Kohonen
  • Computer Science
    Springer Series in Information Sciences
  • 1997

Maximum Likelihood Estimation from Incomplete Data