-
Using autoencoders and deep transfer learning to determine the stellar parameters of 286 CARMENES M dwarfs
Authors:
P. Mas-Buitrago,
A. González-Marcos,
E. Solano,
V. M. Passegger,
M. Cortés-Contreras,
J. Ordieres-Meré,
A. Bello-García,
J. A. Caballero,
A. Schweitzer,
H. M. Tabernero,
D. Montes,
C. Cifuentes
Abstract:
Deep learning (DL) techniques are a promising approach among the set of methods used in the ever-challenging determination of stellar parameters in M dwarfs. In this context, transfer learning could play an important role in mitigating uncertainties in the results due to the synthetic gap (i.e. difference in feature distributions between observed and synthetic data). We propose a feature-based dee…
▽ More
Deep learning (DL) techniques are a promising approach among the set of methods used in the ever-challenging determination of stellar parameters in M dwarfs. In this context, transfer learning could play an important role in mitigating uncertainties in the results due to the synthetic gap (i.e. difference in feature distributions between observed and synthetic data). We propose a feature-based deep transfer learning (DTL) approach based on autoencoders to determine stellar parameters from high-resolution spectra. Using this methodology, we provide new estimations for the effective temperature, surface gravity, metallicity, and projected rotational velocity for 286 M dwarfs observed by the CARMENES survey. Using autoencoder architectures, we projected synthetic PHOENIX-ACES spectra and observed CARMENES spectra onto a new feature space of lower dimensionality in which the differences between the two domains are reduced. We used this low-dimensional new feature space as input for a convolutional neural network to obtain the stellar parameter determinations. We performed an extensive analysis of our estimated stellar parameters, ranging from 3050 to 4300 K, 4.7 to 5.1 dex, and -0.53 to 0.25 dex for Teff, logg, and [Fe/H], respectively. Our results are broadly consistent with those of recent studies using CARMENES data, with a systematic deviation in our Teff scale towards hotter values for estimations above 3750 K. Furthermore, our methodology mitigates the deviations in metallicity found in previous DL techniques due to the synthetic gap. We consolidated a DTL-based methodology to determine stellar parameters in M dwarfs from synthetic spectra, with no need for high-quality measurements involved in the knowledge transfer. These results suggest the great potential of DTL to mitigate the differences in feature distributions between the observations and the PHOENIX-ACES spectra.
△ Less
Submitted 14 May, 2024;
originally announced May 2024.
-
The CARMENES search for exoplanets around M dwarfs -- A deep transfer learning method to determine Teff and [M/H] of target stars
Authors:
A. Bello-García,
V. M. Passegger,
J. Ordieres-Meré,
A. Schweitzer,
J. A. Caballero,
A. González-Marcos,
I. Ribas,
A. Reiners,
A. Quirrenbach,
P. J. Amado,
V. J. S. Béjar,
C. Cifuentes,
Th. Henning,
A. Kaminski,
R. Luque,
D. Montes,
J. C. Morales,
S. Pedraz,
H. M. Tabernero,
M. Zechmeister
Abstract:
The large amounts of astrophysical data being provided by existing and future instrumentation require efficient and fast analysis tools. Transfer learning is a new technique promising higher accuracy in the derived data products, with information from one domain being transferred to improve the accuracy of a neural network model in another domain. In this work, we demonstrate the feasibility of ap…
▽ More
The large amounts of astrophysical data being provided by existing and future instrumentation require efficient and fast analysis tools. Transfer learning is a new technique promising higher accuracy in the derived data products, with information from one domain being transferred to improve the accuracy of a neural network model in another domain. In this work, we demonstrate the feasibility of applying the deep transfer learning (DTL) approach to high-resolution spectra in the framework of photospheric stellar parameter determination. To this end, we used 14 stars of the CARMENES survey sample with interferometric angular diameters to calculate the effective temperature, as well as six M dwarfs that are common proper motion companions to FGK-type primaries with known metallicity. After training a deep learning (DL) neural network model on synthetic PHOENIX-ACES spectra, we used the internal feature representations together with those 14+6 stars with independent parameter measurements as a new input for the transfer process. We compare the derived stellar parameters of a small sample of M dwarfs kept out of the training phase with results from other methods in the literature. Assuming that temperatures from bolometric luminosities and interferometric radii and metallicities from FGK+M binaries are sufficiently accurate, DTL provides a higher accuracy than our previous state-of-the-art DL method (mean absolute differences improve by 20 K for temperature and 0.2 dex for metallicity from DL to DTL when compared with reference values from interferometry and FGK+M binaries). Furthermore, the machine learning (internal) precision of DTL also improves as uncertainties are five times smaller on average. These results indicate that DTL is a robust tool for obtaining M-dwarf stellar parameters comparable to those obtained from independent estimations for well-known stars.
△ Less
Submitted 1 April, 2023;
originally announced April 2023.
-
J-PLUS: Discovery and characterisation of ultracool dwarfs using Virtual Observatory tools II. Second data release and machine learning methodology
Authors:
P. Mas-Buitrago,
E. Solano,
A. González-Marcos,
C. Rodrigo,
E. L. Martín,
J. A. Caballero,
F. Jiménez-Esteban,
P. Cruz,
A. Ederoclite,
J. Ordieres-Meré,
A. Bello-García,
R. A. Dupke,
A. J. Cenarro,
D. Cristóbal-Hornillos,
C. Hernández-Monteagudo,
C. López-Sanjuan,
A. Marín-Franch,
M. Moles,
J. Varela,
H. Vázquez Ramió,
J. Alcaniz,
L. Sodré Jr.,
R. E. Angulo
Abstract:
Ultracool dwarfs (UCDs) comprise the lowest mass members of the stellar population and brown dwarfs, from M7 V to cooler objects with L, T, and Y spectral types. Most of them have been discovered using wide-field imaging surveys, for which the Virtual Observatory (VO) has proven to be of great utility. We aim to perform a search for UCDs in the entire Javalambre Photometric Local Universe Survey (…
▽ More
Ultracool dwarfs (UCDs) comprise the lowest mass members of the stellar population and brown dwarfs, from M7 V to cooler objects with L, T, and Y spectral types. Most of them have been discovered using wide-field imaging surveys, for which the Virtual Observatory (VO) has proven to be of great utility. We aim to perform a search for UCDs in the entire Javalambre Photometric Local Universe Survey (J-PLUS) second data release (2176 deg$^2$) following a VO methodology. We also explore the ability to reproduce this search with a purely machine learning (ML)-based methodology that relies solely on J-PLUS photometry. We followed three different approaches based on parallaxes, proper motions, and colours, respectively, using the VOSA tool to estimate the effective temperatures. For the ML methodology, we built a two-step method based on principal component analysis and support vector machine algorithms. We identified a total of 7827 new candidate UCDs, which represents an increase of about 135% in the number of UCDs reported in the sky coverage of the J-PLUS second data release. Among the candidate UCDs, we found 122 possible unresolved binary systems, 78 wide multiple systems, and 48 objects with a high Bayesian probability of belonging to a young association. We also identified four objects with strong excess in the filter corresponding to the Ca II H and K emission lines and four other objects with excess emission in the H$α$ filter. With the ML approach, we obtained a recall score of 92% and 91% in the test and blind test, respectively. We consolidated the proposed search methodology for UCDs, which will be used in deeper and larger upcoming surveys such as J-PAS and Euclid. We concluded that the ML methodology is more efficient in the sense that it allows for a larger number of true negatives to be discarded prior to analysis with VOSA, although it is more photometrically restrictive.
△ Less
Submitted 19 August, 2022;
originally announced August 2022.
-
Metallicities in M dwarfs: Investigating different determination techniques
Authors:
V. M. Passegger,
A. Bello-García,
J. Ordieres-Meré,
A. Antoniadis-Karnavas,
E. Marfil,
C. Duque-Arribas,
P. J. Amado,
E. Delgado-Mena,
D. Montes,
B. Rojas-Ayala,
A. Schweitzer,
H. M. Tabernero,
V. J. S. Béjar,
J. A. Caballero,
A. P. Hatzes,
Th. Henning,
S. Pedraz,
A. Quirrenbach,
A. Reiners,
I. Ribas
Abstract:
Deriving metallicities for solar-like stars follows well-established methods, but for cooler stars such as M dwarfs, the determination is much more complicated due to forests of molecular lines that are present. Several methods have been developed in recent years to determine accurate stellar parameters for these cool stars ($T_{\rm eff} \lesssim$ 4000 K). However, significant differences can be f…
▽ More
Deriving metallicities for solar-like stars follows well-established methods, but for cooler stars such as M dwarfs, the determination is much more complicated due to forests of molecular lines that are present. Several methods have been developed in recent years to determine accurate stellar parameters for these cool stars ($T_{\rm eff} \lesssim$ 4000 K). However, significant differences can be found at times when comparing metallicities for the same star derived using different methods. In this work, we determine the effective temperatures, surface gravities, and metallicities of 18 well-studied M dwarfs observed with the CARMENES high-resolution spectrograph following different approaches, including synthetic spectral fitting, analysis of pseudo-equivalent widths, and machine learning. We analyzed the discrepancies in the derived stellar parameters, including metallicity, in several analysis runs. Our goal is to minimize these discrepancies and find stellar parameters that are more consistent with the literature values. We attempted to achieve this consistency by standardizing the most commonly used components, such as wavelength ranges, synthetic model spectra, continuum normalization methods, and stellar parameters. We conclude that although such modifications work quite well for hotter main-sequence stars, they do not improve the consistency in stellar parameters for M dwarfs, leading to mean deviations of around 50-200 K in temperature and 0.1-0.3 dex in metallicity. In particular, M dwarfs are much more complex and a standardization of the aforementioned components cannot be considered as a straightforward recipe for bringing consistency to the derived parameters. Further in-depth investigations of the employed methods would be necessary in order to identify and correct for the discrepancies that remain.
△ Less
Submitted 29 November, 2021;
originally announced November 2021.
-
The CARMENES search for exoplanets around M dwarfs -- A deep learning approach to determine fundamental parameters of target stars
Authors:
V. M. Passegger,
A. Bello-García,
J. Ordieres-Meré,
J. A. Caballero,
A. Schweitzer,
A. González-Marcos,
I. Ribas,
A. Reiners,
A. Quirrenbach,
P. J. Amado,
M. Azzaro,
F. F. Bauer,
V. J. S. Béjar,
M. Cortés-Contreras,
S. Dreizler,
A. P. Hatzes,
Th. Henning,
S. V. Jeffers,
A. Kaminski,
M. Kürster,
M. Lafarga,
E. Marfil,
D. Montes,
J. C. Morales,
E. Nagel
, et al. (4 additional authors not shown)
Abstract:
Existing and upcoming instrumentation is collecting large amounts of astrophysical data, which require efficient and fast analysis techniques. We present a deep neural network architecture to analyze high-resolution stellar spectra and predict stellar parameters such as effective temperature, surface gravity, metallicity, and rotational velocity. With this study, we firstly demonstrate the capabil…
▽ More
Existing and upcoming instrumentation is collecting large amounts of astrophysical data, which require efficient and fast analysis techniques. We present a deep neural network architecture to analyze high-resolution stellar spectra and predict stellar parameters such as effective temperature, surface gravity, metallicity, and rotational velocity. With this study, we firstly demonstrate the capability of deep neural networks to precisely recover stellar parameters from a synthetic training set. Secondly, we analyze the application of this method to observed spectra and the impact of the synthetic gap (i.e., the difference between observed and synthetic spectra) on the estimation of stellar parameters, their errors, and their precision. Our convolutional network is trained on synthetic PHOENIX-ACES spectra in different optical and near-infrared wavelength regions. For each of the four stellar parameters, $T_{\rm eff}$, $\log{g}$, [M/H], and $v \sin{i}$, we constructed a neural network model to estimate each parameter independently. We then applied this method to 50 M dwarfs with high-resolution spectra taken with CARMENES (Calar Alto high-Resolution search for M dwarfs with Exo-earths with Near-infrared and optical Echelle Spectrographs), which operates in the visible (520-960 nm) and near-infrared wavelength range (960-1710 nm) simultaneously. Our results are compared with literature values for these stars. They show mostly good agreement within the errors, but also exhibit large deviations in some cases, especially for [M/H], pointing out the importance of a better understanding of the synthetic gap.
△ Less
Submitted 3 August, 2020;
originally announced August 2020.