Using transfer learning to detect galaxy mergers

@article{Ackermann2018UsingTL,
  title={Using transfer learning to detect galaxy mergers},
  author={Sandro Ackermann and Kevin Schawinski and Ce Zhang and Anna K. Weigel and M. Dennis Turp},
  journal={ArXiv},
  year={2018},
  volume={abs/1805.10289}
}
We investigate the use of deep convolutional neural networks (deep CNNs) for automatic visual detection of galaxy mergers. Moreover, we investigate the use of transfer learning in conjunction with CNNs, by retraining networks first trained on pictures of everyday objects. We test the hypothesis that transfer learning is useful for improving classification performance for small training sets. This would make transfer learning useful for finding rare objects in astronomical imaging datasets. We… 
Identifying galaxy mergers in observations and simulations with deep learning
TLDR
Classifying SDSS observations with the simulation trained network has proven to work, providing tantalising prospects for using simulation trained networks for galaxy identification in large surveys and suggests that most of the simulated mergers do not have conspicuous merger features and visually identified merger catalogues from observations are incomplete and biased towards certain merger types.
DeepMerge: Classifying High-redshift Merging Galaxies with Deep Neural Networks
DeepMerge II: Building Robust Deep Learning Algorithms for Merging Galaxy Identification Across Domains
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
Deep learning predictions of galaxy merger stage and the importance of observational realism
Machine learning is becoming a popular tool to quantify galaxy morphologies and identify mergers. However, this technique relies on using an appropriate set of training data to be successful. By
Convolutional neural network identification of galaxy post-mergers in UNIONS using IllustrisTNG
The Canada-France Imaging Survey (CFIS) will consist of deep, high-resolution r-band imaging over ~5000 square degrees of the sky, representing a first-rate opportunity to identify recently-merged
Classifying the formation processes of S0 galaxies using Convolutional Neural Networks
Numerous studies have demonstrated the ability of Convolutional Neural Networks (CNNs) to classify large numbers of galaxies in a manner that mimics the expertise of astronomers. Such classifications
Domain adaptation techniques for improved cross-domain study of galaxy mergers
TLDR
It is shown how the addition of either MMD or adversarial training greatly improves the performance of the classifier on the target domain when compared to conventional machine learning algorithms, thereby demonstrating great promise for their use in astronomy.
Morphological classification of compact and extended radio galaxies using convolutional neural networks and data augmentation techniques
TLDR
Results show that model developed in this work is able to identify different morphological classes of radio galaxies with a high efficiency and performance.
Deep transfer learning for star cluster classification: I. application to the PHANGS–HST survey
TLDR
Given the relatively small and unbalanced nature of existing, human-labelled star cluster datasets, the knowledge of neural network models for real-object recognition is transferred to classify star clusters candidates into four morphological classes to automate classification for these objects at scale, and the creation of a standardized dataset.
...
...

References

SHOWING 1-10 OF 59 REFERENCES
Rotation-invariant convolutional neural networks for galaxy morphology prediction
TLDR
A deep neural network model for galaxy morphology classification which exploits translational and rotational symmetry is developed in the context of the Galaxy Challenge, an international competition to build the best model for morphology classification based on annotated images from the Galaxy Zoo project.
ImageNet classification with deep convolutional neural networks
TLDR
A large, deep convolutional neural network was trained to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes and employed a recently developed regularization method called "dropout" that proved to be very effective.
Measuring photometric redshifts using galaxy images and Deep Neural Networks
  • B. Hoyle
  • Computer Science, Physics
    Astron. Comput.
  • 2016
Xception: Deep Learning with Depthwise Separable Convolutions
  • François Chollet
  • Computer Science
    2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2017
TLDR
This work proposes a novel deep convolutional neural network architecture inspired by Inception, where Inception modules have been replaced with depthwise separable convolutions, and shows that this architecture, dubbed Xception, slightly outperforms Inception V3 on the ImageNet dataset, and significantly outperforms it on a larger image classification dataset.
Deep Learning
TLDR
Deep learning is making major advances in solving problems that have resisted the best attempts of the artificial intelligence community for many years, and will have many more successes in the near future because it requires very little engineering by hand and can easily take advantage of increases in the amount of available computation and data.
A new automatic method to identify galaxy mergers – I. Description and application to the Space Telescope A901/902 Galaxy Evolution Survey★
We present a new automatic method to identify galaxy mergers using the morphological information contained in the residual images of galaxies after the subtraction of a smooth Sersic model. The
Space Warps – I. Crowdsourcing the discovery of gravitational lenses
TLDR
Comment on the scalability of the SpaceWarps system to the wide field survey era, based on the projection that searches of 105 images could be performed by a crowd of 105 volunteers in 6 days.
A NEW NONPARAMETRIC APPROACH TO GALAXY MORPHOLOGICAL CLASSIFICATION
We present two new nonparametric methods for quantifying galaxy morphology: the relative distribution of the galaxy pixel flux values (the Gini coefficient or G) and the second-order moment of the
Galaxy Zoo: morphologies derived from visual inspection of galaxies from the Sloan Digital Sky Survey
In order to understand the formation and subsequent evolution of galaxies one must first distinguish between the two main morphological classes of massive systems: spirals and early-type systems.
Galaxy Zoo: major galaxy mergers are not a significant quenching pathway
We use stellar mass functions to study the properties and the significance of quenching through major galaxy mergers. In addition to SDSS DR7 and Galaxy Zoo 1 data, we use samples of visually
...
...