Bayesian Deep Learning for Exoplanet Atmospheric Retrieval. (arXiv:1811.03390v1 [astro-ph.EP])
<a href="http://arxiv.org/find/astro-ph/1/au:+Soboczenski_F/0/1/0/all/0/1">Frank Soboczenski</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Himes_M/0/1/0/all/0/1">Michael D. Himes</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+OBeirne_M/0/1/0/all/0/1">Molly D. O'Beirne</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Zorzan_S/0/1/0/all/0/1">Simone Zorzan</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Baydin_A/0/1/0/all/0/1">Atilim Gunes Baydin</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Cobb_A/0/1/0/all/0/1">Adam D. Cobb</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Angerhausen_D/0/1/0/all/0/1">Daniel Angerhausen</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Arney_G/0/1/0/all/0/1">Giada N. Arney</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Domagal_Goldman_S/0/1/0/all/0/1">Shawn D. Domagal-Goldman</a>
Over the past decade, the study of exoplanets has shifted from their
detection to the characterization of their atmospheres. Atmospheric retrieval,
the inverse modeling technique used to determine an atmosphere’s temperature
and composition from an observed spectrum, is both time-consuming and
compute-intensive, requiring complex algorithms that compare thousands to
millions of atmospheric models to the observational data to find the most
probable values and associated uncertainties for each model parameter. For
rocky, terrestrial planets, the retrieved atmospheric composition can give
insight into the surface fluxes of gaseous species necessary to maintain the
stability of that atmosphere, which may in turn provide insight into the
geological and/or biological processes active on the planet. These atmospheres
contain many molecules, some of which are biosignatures, or molecules
indicative of biological activity. Runtimes of traditional retrieval models
scale with the number of model parameters, so as more molecular species are
considered, runtimes can become prohibitively long. Recent advances in machine
learning (ML) and computer vision offer new ways to reduce the time to perform
a retrieval by orders of magnitude, given a sufficient data set to train with.
Here we present an ML-based retrieval framework called Intelligent exoplaNet
Atmospheric RetrievAl (INARA) that consists of a Bayesian deep learning model
for retrieval and a data set of 3,000,000 spectra of synthetic rocky exoplanets
generated using the NASA Planetary Spectrum Generator (PSG). Our work
represents the first ML model for rocky, terrestrial exoplanets and the first
synthetic data set of spectra generated at this scale.
Over the past decade, the study of exoplanets has shifted from their
detection to the characterization of their atmospheres. Atmospheric retrieval,
the inverse modeling technique used to determine an atmosphere’s temperature
and composition from an observed spectrum, is both time-consuming and
compute-intensive, requiring complex algorithms that compare thousands to
millions of atmospheric models to the observational data to find the most
probable values and associated uncertainties for each model parameter. For
rocky, terrestrial planets, the retrieved atmospheric composition can give
insight into the surface fluxes of gaseous species necessary to maintain the
stability of that atmosphere, which may in turn provide insight into the
geological and/or biological processes active on the planet. These atmospheres
contain many molecules, some of which are biosignatures, or molecules
indicative of biological activity. Runtimes of traditional retrieval models
scale with the number of model parameters, so as more molecular species are
considered, runtimes can become prohibitively long. Recent advances in machine
learning (ML) and computer vision offer new ways to reduce the time to perform
a retrieval by orders of magnitude, given a sufficient data set to train with.
Here we present an ML-based retrieval framework called Intelligent exoplaNet
Atmospheric RetrievAl (INARA) that consists of a Bayesian deep learning model
for retrieval and a data set of 3,000,000 spectra of synthetic rocky exoplanets
generated using the NASA Planetary Spectrum Generator (PSG). Our work
represents the first ML model for rocky, terrestrial exoplanets and the first
synthetic data set of spectra generated at this scale.
http://arxiv.org/icons/sfx.gif
