Application of convolutional neural networks for stellar spectral classification

Abstract

Due to the ever-expanding volume of observed spectroscopic data from surveys such as SDSS and LAMOST, it has become important to apply artificial intelligence (AI) techniques for analysing stellar spectra to solve spectral classification and regression problems like the determination of stellar atmospheric parameters T eff , log g, and [Fe/H]. We propose an automated approach for the classification of stellar spectra in the optical region using Convolutional Neural Networks. Traditional machine learning (ML) methods with "shallow" architecture (usually up to 2 hidden layers) have been trained for these purposes in the past. However, deep learning methods with a larger number of hidden layers allow the use of finer details in the spectrum which results in improved accuracy and better generalisation. Studying finer spectral signatures also enables us to determine accurate differential stellar parameters and find rare objects. We examine various machine and deep learning algorithms like Artificial Neural Networks (ANN), Random Forest (RF), and Convolutional Neural Network (CNN) to classify stellar spectra using the Jacoby Atlas, ELODIE and MILES spectral libraries as training samples. We test the performance of the trained networks on the Indo-U.S. Library of Coudé Feed Stellar Spectra (CFLIB). We show that using convolutional neural networks, we are able to lower the error up to 1.23 spectral sub-classes as compared to that of 2 sub-classes achieved in the past studies with ML approach. We further apply the trained model to classify stellar spectra retrieved from the SDSS database with SNR>20.

The CARMENES search for exoplanets around M dwarfs

“…Different DL architectures were considered, some of them inspired by literature such as Sharma et al (2019), StarNet (Kielty et al 2018), and many other homemade architectures. To this end, a flexible python code was implemented where the topology for the convolutional structure and for the ANN layers were passed as parameters.…”

“…Early applications of neural networks to characterize stellar spectra can be found in von Hippel et al (1994), Gulati et al (1994), andSingh et al (1998), for example. Bailer- Jones et al (1997) trained an artificial neural network with synthetic spectra to determine T eff , log g, and [M/H] for over 5000 stars of spectral types B to K. Sharma et al (2019) compared different ML algorithms, such as artificial neural networks (ANN), random forests, and convolutional neural networks, to classify stellar spectra. They report that their convolutional neural network achieved better accuracy than the other ML algorithms, and point out the importance of a sufficiently large training set.…”

A Blended Artificial Intelligence Approach for Spectral Classification of Stars in Massive Astronomical Surveys

“…Certain well-known previous works have also applied AI techniques and deep learning to the stellar classification field [ 6 , 7 , 8 , 9 , 10 , 11 , 12 ], obtaining different performance in the classification. The big data and AI processing on massive catalogs (that do not stop growing in number) go hand in hand.…”

Prediction of parameters and classification of molecular outflows using convolutional neural networks

“…Крупные наблюдательные обзоры предоставили массивные наборы данных для разработки инструментов машинного обучения для решения прикладных задач астрофизики, что сделало машинное обучение еще более привлекательным [1]. Методы машинного обучения были применены для классификации звезд/галактик и определении физических параметров [2][3][4][5][6][7][8][9][10][11]. Еще одним успехом ML в астрофизике стало использование архитектуры глубокой нейронной сети для анализа звездных спектров [12].…”