Transfer learning for radio galaxy classification

Abstract

In the context of radio galaxy classification, most state-of-the-art neural network algorithms have been focused on single survey data. The question of whether these trained algorithms have cross-survey identification ability or can be adapted to develop classification networks for future surveys is still unclear. One possible solution to address this issue is transfer learning, which re-uses elements of existing machine learning models for different applications. Here we present radio galaxy classification based on a 13-layer Deep Convolutional Neural Network (DCNN) using transfer learning methods between different radio surveys. We find that our machine learning models trained from a random initialization achieve accuracies comparable to those found elsewhere in the literature. When using transfer learning methods, we find that inheriting model weights pre-trained on FIRST images can boost model performance when re-training on lower resolution NVSS data, but that inheriting pre-trained model weights from NVSS and re-training on FIRST data impairs the performance of the classifier. We consider the implication of these results in the context of future radio surveys planned for next-generation radio telescopes such as ASKAP, MeerKAT, and SKA1-MID.

References in corpus (29)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• An overview of gradient descent optimization algorithms
• The Origin of the Mass--Metallicity Relation: Insights from 53,000 Star-Forming Galaxies in the SDSS
• The physical properties of star forming galaxies in the low redshift universe
• Planck 2018 results. I. Overview and the cosmological legacy of Planck
• SUMSS: A Wide-Field Radio Imaging Survey of the Southern Sky. II. The Source Catalogue
• Visualizing and Understanding Convolutional Networks
• EMU: Evolutionary Map of the Universe
• Giant Radio Sources
• Radio Galaxy Zoo: host galaxies and radio morphologies derived from visual inspection
• Using transfer learning to detect galaxy mergers
• Radio Galaxy Zoo: ClaRAN - A Deep Learning Classifier for Radio Morphologies
• On the deceleration of FRI jets: mass loading by stellar-winds
• Transfer learning for galaxy morphology from one survey to another
• Radio Galaxy Zoo: Compact and extended radio source classification with deep learning
• Applying Deep Learning to Fast Radio Burst Classification
• Training Deep Neural Networks via Direct Loss Minimization
• Discovery of Giant Radio Galaxies from NVSS: Radio & Infrared Properties
• ASKAP HI imaging of the galaxy group IC 1459
• Radio Galaxy Zoo: Machine learning for radio source host galaxy cross-identification
• The CoNFIG Catalogue - II. Comparison of Space Densities in the FR Dichotomy
• Video-based Sign Language Recognition without Temporal Segmentation
• Understanding the Disharmony between Dropout and Batch Normalization by Variance Shift
• Deep Learning Based Detection of Cosmological Diffuse Radio Sources
• Optical and radio properties of giant radio quasars: Central black hole characteristics
• A Deep Architecture for Semantic Parsing
• Fine-tuning deep CNN models on specific MS COCO categories

Cited by in corpus (24)

• Morphological classification of radio galaxies: Capsule Networks versus Convolutional Neural Networks
• Practical Galaxy Morphology Tools from Deep Supervised Representation Learning
• Cataloging the radio-sky with unsupervised machine learning: a new approach for the SKA era
• CNN Architecture Comparison for Radio Galaxy Classification
• Fanaroff-Riley classification of radio galaxies using group-equivariant convolutional neural networks
• Morphological classification of compact and extended radio galaxies using convolutional neural networks and data augmentation techniques
• Learning to detect RFI in radio astronomy without seeing it
• The GLEAM 4-Jy (G4Jy) Sample: II. Host-galaxy identification for individual sources
• Deep Learning for Cardiologist-level Myocardial Infarction Detection in Electrocardiograms
• SKA Science Data Challenge 2: analysis and results
• A machine learning classifier for LOFAR radio galaxy cross-matching techniques
• Quantifying Uncertainty in Deep Learning Approaches to Radio Galaxy Classification
• Structured Variational Inference for Simulating Populations of Radio Galaxies
• Morphological Classification of Radio Galaxies with wGAN-supported Augmentation
• Subaru Hyper Suprime-Cam revisits the large-scale environmental dependence on galaxy morphology over 360 deg${}^2$ at z=0.3-0.6
• Deep Transfer Learning for Classification of Variable Sources
• Identification of 4876 Bent-Tail Radio Galaxies in the FIRST Survey Using Deep Learning Combined with Visual Inspection
• Automated classification of Chandra X-ray point sources using machine learning methods
• Radio Galaxy Zoo: Using semi-supervised learning to leverage large unlabelled data-sets for radio galaxy classification under data-set shift
• Radio Halo Detection in MWA Data using Deep Neural Networks and Generative Data Augmentation
• Radio Galaxy Zoo: Giant Radio Galaxy Classification using Multi-Domain Deep Learning
• E(2) Equivariant Self-Attention for Radio Astronomy
• Radio Galaxy Zoo: Morphological classification by Fanaroff-Riley designation using self-supervised pre-training
• Can semi-supervised learning reduce the amount of manual labelling required for effective radio galaxy morphology classification?