J/MNRAS/469/4578 Deep learning classification in asteroseismology (Hon+, 2017)
Deep learning classification in asteroseismology.
Hon M., Stello D., Yu J.
<Mon. Not. R. Astron. Soc., 469, 4578-4583 (2017)>
=2017MNRAS.469.4578H 2017MNRAS.469.4578H (SIMBAD/NED BibCode)
ADC_Keywords: Asteroseismology ; Stars, giant ; Models
Keywords: asteroseismology - methods: data analysis -
techniques: image processing - stars: oscillations - stars: statistics
Abstract:
In the power spectra of oscillating red giants, there are visually
distinct features defining stars ascending the red giant branch from
those that have commenced helium core burning. We train a 1D
convolutional neural network by supervised learning to automatically
learn these visual features from images of folded oscillation spectra.
By training and testing on Kepler red giants, we achieve an accuracy
of up to 99 per cent in separating helium-burning red giants from
those ascending the red giant branch. The convolutional neural network
additionally shows capability in accurately predicting the
evolutionary states of 5379 previously unclassified Kepler red giants,
by which we now have greatly increased the number of classified stars.
Description:
We obtain the evolutionary state classifications of 5673 Kepler stars
based on automated asymptotic period spacing measurements by Vrard et
al. (2016A&A...588A..87V 2016A&A...588A..87V, Cat. J/A+A/588/A87, hereafter V16), and add
335 stars from the classification by Mosser et al.
(2014A&A...572L...5M 2014A&A...572L...5M, Cat. J/A+A/572/L5, hereafter M14) that are not
already in V16's sample, to a total of 6008 stars. We then assign RGB
stars with the binary class 0 and HeB stars with class 1. About 30 per
cent of stars in the data set are RGB stars. We randomly choose 1008
stars as test data, with the remaining 5000 stars for training.
Additionally, we have an unclassified set comprising 8794 Kepler red
giants that are known to oscillate but have not been given
classifications by V16 or M14. We want to predict the population
labels of all stars in our unclassified set using our trained neural
network
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
test.dat 54 1008 *Population class predictions for 1008 Kepler
red giants (test set)
unclass.dat 55 7655 *Population class predictions for 7655 Kepler
red giants (the unclassified set)
--------------------------------------------------------------------------------
Note on *.dat: from the deep learning classifier.
--------------------------------------------------------------------------------
See also:
V/133 : Kepler Input Catalog (Kepler Mission Team, 2009)
Byte-by-byte Description of file: test.dat unclass.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 8 I8 --- KIC KIC number
10 A1 --- Pop [0/1] Population (1)
16- 23 F8.5 uHz Dnu Mean large frequency separation of modes with
the same degree and consecutive order,
{DELTA}nu
28- 36 F9.5 uHz numax Frequency of maximum oscillation power
43- 49 F7.3 --- epsilon ?=-99 Location of the l=0 mode (2)
53- 55 A3 --- Com [ESNVM] Comments (3)
--------------------------------------------------------------------------------
Note (1): Population as follows:
0 = RGB
1 = HeB
Note (2): epsilon=1/4+alpha, where alpha is the contribution from the outer
turning point, which is determined by the properties of the near-surface
region of the star (see Huber et al. (2010ApJ...723.1607H 2010ApJ...723.1607H), Sect. 3.3).
Values of epsilon = -99 indicate that epsilon was unobtainable.
Note (3): Comments as follows:
V = Disputed by Vrard et al. (2016A&A...588A..87V 2016A&A...588A..87V, Cat. J/A+A/588/A87)
M = Disputed by Mosser et al. (2014A&A...572L...5M 2014A&A...572L...5M, Cat. J/A+A/572/L5)
E = Disputed by Elsworth et al. (2017MNRAS.466.3344E 2017MNRAS.466.3344E, Cat. J/MNRAS/466/3344)
S = Disputed by Stello et al. (2013ApJ...765L..41S 2013ApJ...765L..41S, Cat. J/ApJ/765/L41)
N = HeB prediction at numax>110uHz, a range where HeB stars should
not exist. A likely incorrect prediction
--------------------------------------------------------------------------------
History:
From electronic version of the journal
(End) Patricia Vannier [CDS] 19-Apr-2020