Finding flares in Kepler and TESS data with recurrent deep neural networks

Abstract

Stellar flares are an important aspect of magnetic activity -- both for stellar evolution and circumstellar habitability viewpoints - but automatically and accurately finding them is still a challenge to researchers in the Big Data era of astronomy. We present an experiment to detect flares in space-borne photometric data using deep neural networks. Using a set of artificial data and real photometric data we trained a set of neural networks, and found that the best performing architectures were the recurrent neural networks (RNNs) using Long Short-Term Memory (LSTM) layers. The best trained network detected flares over {$5\sigma$ with $\gtrsim 80$\% recall and precision} and was also capable to distinguish typical false signals (e.g. maxima of RR Lyr stars) from real flares. Testing the network trained on Kepler data on TESS light curves showed that the neural net is able to generalize and find flares - with similar effectiveness - in completely new data having previously unseen sampling and characteristics.

arXiv comment9 pages (+1 appendix), 10 figures, accepted for publication at A&A

References in corpus (17)

• SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
• The NumPy array: a structure for efficient numerical computation
• The Transiting Exoplanet Survey Satellite
• The PLATO 2.0 Mission
• Information criteria for astrophysical model selection
• Kepler Input Catalog: Photometric Calibration and Stellar Classification
• Stellar Flares from the First Tess Data Release: Exploring a New Sample of M-dwarfs
• The Kepler Catalog of Stellar Flares
• White-light flares on cool stars in the Kepler Quarter 1 Data
• Frequent flaring in the TRAPPIST-1 system - unsuited for life?
• Flare Statistics for Young Stars from a Convolutional Neural Network Analysis of $\text{TESS}$ Data
• A path towards understanding the rotation-activity relation of M dwarfs with K2 mission, X-ray and UV data
• An Analytic Model for Rotational Modulations in the Photometry of Spotted Stars
• Finding flares in Kepler data using machine learning tools
• The Connection between Starspots and Flares on Main-Sequence Kepler Stars
• Optical flares from the faint mid-dM star 2MASS J00453911+4140396
• Towards the true number of flaring giant stars in the Kepler field. Are there flaring specialities associated with the giant nature?

Cited by in corpus (16)

• Searching for flaring star-planet interactions in AU Mic TESS observations
• Testing Self-Organized Criticality Across the Main Sequence using Stellar Flares from TESS
• Short term variability of DS Tuc A observed with TESS
• Variable star classification with a Multiple-Input Neural Network
• Characteristics of flares on giant stars
• Stellar flare morphology with TESS across the main sequence
• A confined dynamo: magnetic activity of the K-dwarf component in the pre-cataclysmic binary system V471 Tauri
• Convolutional Neural Networks for Searching Superflares from Pixel-level Data of the Transiting Exoplanet Survey Satellite
• Effects of transient stellar emissions on planetary climates of tidally-locked exo-earths
• Deep learning-based astronomical multimodal data fusion: A comprehensive review
• Enhancing the detection of low-energy M dwarf flares: Wavelet-based denoising of CHEOPS data
• FALCO: a Foundation model of Astronomical Light Curves for time dOmain astronomy
• Glancing through the debris disk: Photometric analysis of DE Boo with CHEOPS
• Heartbeat Stars Recognition Based on Recurrent Neural Networks: Method and Validation
• Radio emission from flaring stars and brown dwarfs
• Characterisation of starspot structure and differential rotation of Kepler-411