{"id":37154,"date":"2018-10-10T22:29:24","date_gmt":"2018-10-10T19:29:24","guid":{"rendered":"https:\/\/www.altoros.com\/blog\/?p=37154"},"modified":"2019-09-20T20:48:11","modified_gmt":"2019-09-20T17:48:11","slug":"how-nasa-uses-artificial-intelligence-to-detect-exoplanets","status":"publish","type":"post","link":"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/","title":{"rendered":"How NASA Uses Artificial Intelligence to Detect Exoplanets"},"content":{"rendered":"<div id=\"ez-toc-container\" class=\"ez-toc-v2_0_79_2 counter-hierarchy ez-toc-counter ez-toc-transparent ez-toc-container-direction\">\n<div class=\"ez-toc-title-container\">\n<p class=\"ez-toc-title\" style=\"cursor:inherit\">Table of Contents<\/p>\n<span class=\"ez-toc-title-toggle\"><a href=\"#\" class=\"ez-toc-pull-right ez-toc-btn ez-toc-btn-xs ez-toc-btn-default ez-toc-toggle\" aria-label=\"Toggle Table of Content\"><span class=\"ez-toc-js-icon-con\"><span class=\"\"><span class=\"eztoc-hide\" style=\"display:none;\">Toggle<\/span><span class=\"ez-toc-icon-toggle-span\"><svg style=\"fill: #999;color:#999\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" class=\"list-377408\" width=\"20px\" height=\"20px\" viewBox=\"0 0 24 24\" fill=\"none\"><path d=\"M6 6H4v2h2V6zm14 0H8v2h12V6zM4 11h2v2H4v-2zm16 0H8v2h12v-2zM4 16h2v2H4v-2zm16 0H8v2h12v-2z\" fill=\"currentColor\"><\/path><\/svg><svg style=\"fill: #999;color:#999\" class=\"arrow-unsorted-368013\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" width=\"10px\" height=\"10px\" viewBox=\"0 0 24 24\" version=\"1.2\" baseProfile=\"tiny\"><path d=\"M18.2 9.3l-6.2-6.3-6.2 6.3c-.2.2-.3.4-.3.7s.1.5.3.7c.2.2.4.3.7.3h11c.3 0 .5-.1.7-.3.2-.2.3-.5.3-.7s-.1-.5-.3-.7zM5.8 14.7l6.2 6.3 6.2-6.3c.2-.2.3-.5.3-.7s-.1-.5-.3-.7c-.2-.2-.4-.3-.7-.3h-11c-.3 0-.5.1-.7.3-.2.2-.3.5-.3.7s.1.5.3.7z\"\/><\/svg><\/span><\/span><\/span><\/a><\/span><\/div>\n<nav><ul class='ez-toc-list ez-toc-list-level-1 ' ><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-1\" href=\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#Houston_I_want_to_believe\" >Houston, I want to believe!<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-2\" href=\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#Using_transit_methods_to_detect_an_exoplanet\" >Using transit methods to detect an exoplanet<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-3\" href=\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#Planetary_Spectrum_Generator\" >Planetary Spectrum Generator<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-4\" href=\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#INARA_Intelligent_Exoplanet_Atmosphere_Retrieval\" >INARA: Intelligent Exoplanet Atmosphere Retrieval<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-5\" href=\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#Want_details_Watch_the_video\" >Want details? Watch the video!<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-6\" href=\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#Related_slides\" >Related slides<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-7\" href=\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#Further_reading\" >Further reading<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-8\" href=\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#About_the_expert\" >About the expert<\/a><\/li><\/ul><\/nav><\/div>\n<h3><span class=\"ez-toc-section\" id=\"Houston_I_want_to_believe\"><\/span>Houston, I want to believe!<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<p>Being just a tiny blue dot in the ever-expanding universe, humans of Earth have been wondering for centuries if they are alone here. The answer to this question is the motivator behind scientific research of the world&#8217;s best minds, exquisitely equipped observatories, neatly designed spacecrafts, and the rapidly developing and emerging technologies.<\/p>\n<p>To get the answer, other planets, naturally, first have to be discovered. Any planet is just a faint source of light in comparison to the parent star it is orbiting. For instance, the Sun is a billion times as bright as the reflected light from Earth or Mars, or any of the other planets in our solar system. This gives astronomers a double trouble of identifying an exoplanet that provides a hard-to-distinguish source of light, which is outwashed by a powerful glare of the parent star at the same time.<\/p>\n<div id=\"attachment_37227\" style=\"width: 160px\" class=\"wp-caption alignright\"><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/William-Fawcett-NASA-Frontier-Development-Lab.jpg\"><img loading=\"lazy\" decoding=\"async\" aria-describedby=\"caption-attachment-37227\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/William-Fawcett-NASA-Frontier-Development-Lab-150x150.jpg\" alt=\"\" width=\"150\" height=\"150\" class=\"size-thumbnail wp-image-37227\" \/><\/a><p id=\"caption-attachment-37227\" class=\"wp-caption-text\"><small>William Fawcett<\/small><\/p><\/div>\n<p>Due to this reasons, very few extrasolar planets can be observed directly, and even fewer can be distinguished from its parent star. As of September 27, 2018, there were <a href=\"https:\/\/exoplanetarchive.ipac.caltech.edu\/\" rel=\"noopener noreferrer\" target=\"_blank\">3,791 confirmed exoplanets<\/a>, most of which were detected by the <a href=\"https:\/\/www.nasa.gov\/ames\/kepler\/kepler-spacecraft-updates\/\" rel=\"noopener noreferrer\" target=\"_blank\">Kepler<\/a> spacecraft. For instance, there were 1,300 exoplanets <a href=\"https:\/\/en.wikipedia.org\/wiki\/List_of_exoplanets_discovered_in_2016\" rel=\"noopener noreferrer\" target=\"_blank\">found<\/a> in 2016.<\/p>\n<p>At the recent <a href=\"https:\/\/www.meetup.com\/ru-RU\/TensorFlow-London\/events\/254471478\/\" rel=\"noopener noreferrer\" target=\"_blank\">TensorFlow meetup<\/a> in London, <a href=\"https:\/\/www.linkedin.com\/in\/william-fawcett\/\" rel=\"noopener noreferrer\" target=\"_blank\">William Fawcett<\/a> of the NASA Frontier Development Lab shared insights to how the institution uses artificial intelligence to find life beyond Earth.<\/p>\n<blockquote><p><em>&#8220;The galaxy is very big, I&#8217;m sure you&#8217;re all aware. There is something like 100 billion stars, and we think about 40 billion of those stars have a planet orbiting it that could potentially host life.&#8221; \u2014William Fawcett, NASA Frontier Development Lab<\/em><\/p><\/blockquote>\n<p>&nbsp;<\/p>\n<h3><span class=\"ez-toc-section\" id=\"Using_transit_methods_to_detect_an_exoplanet\"><\/span>Using transit methods to detect an exoplanet <span class=\"ez-toc-section-end\"><\/span><\/h3>\n<p>Exoplanets can be detected using the <a href=\"https:\/\/en.wikipedia.org\/wiki\/Methods_of_detecting_exoplanets\" rel=\"noopener noreferrer\" target=\"_blank\">transit techniques<\/a>, which imply measuring the brightness of a target star as a function of time, producing a flux time series called a light curve. Simply put, exoplanets are detected when they transit in front of a star like our Sun and cause a drop in the measured brightness.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-the-transit-techniques-to-detect-exoplanets-artificial-intelligence-machine-learning.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-the-transit-techniques-to-detect-exoplanets-artificial-intelligence-machine-learning.png\" alt=\"\" width=\"640\" class=\"aligncenter size-full wp-image-37243\" \/><\/a><small>Stars transiting in front of the Sun (<a href=\"https:\/\/www.slideshare.net\/seldon_io\/tensorflow-london-17-how-nasa-frontier-development-lab-scientists-use-ai-to-find-evidence-of-life\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>However, the signals of such a star passing are hardly traceable in comparison to the instrumental noise and systematics, as well as the inherent stellar variation present in the data captured and measured. In addition, such false-positive planet signals as background eclipsing binaries, for example, should be removed to achieve a reliable result. With all the massive data sets of light curves produced by Kepler and other spacecrafts sailing the universe, scientists need efficient means of processing and analyzing the data. This is where machine learning comes helpful.<\/p>\n<blockquote><p><em>&#8220;Detecting an exoplanet is as tricky to spot as a firefly flying next to a searchlight from thousands of miles away.&#8221; \u2014William Fawcett, NASA Frontier Development Lab<\/em><\/p><\/blockquote>\n<p>Detecting an exoplanet is still just a beginning of the mission. Now, the scientists have to find some evidence of life to label the planet as the one potentially hosting life. The next step is <em>measuring atmospheric spectrum<\/em>, which is essential to understanding what a planet&#8217;s size and atmospheric properties are. The method underlying such findings is <a href=\"https:\/\/arxiv.org\/pdf\/1709.05941.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">transmission spectroscopy<\/a>.<\/p>\n<p>When a planet passes the parent star, it blocks a fraction of the stellar flux equal to the sky-projected area of the planet relative to the area of the star. The fractional drop in flux is referred to as the transit depth. The main idea behind transmission spectroscopy is that the planet\u2019s transit depth is wavelength-dependent. At wavelengths where the atmosphere is more opaque due to the absorption by atoms or molecules, the planet blocks a bit more of stellar flux. To measure these variations, the light curve is binned in wavelength into spectrophotometric channels, and the light curve from each channel is fit separately with a transit model. The measured transit depths as a function of wavelength constitute the transmission spectrum.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-artificial-intelligence-measuring-atmospheric-spectrum.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-artificial-intelligence-measuring-atmospheric-spectrum.png\" alt=\"\" width=\"640\" class=\"aligncenter size-full wp-image-37275\" \/><\/a><small>Using transmission spectroscopy to measure atmospheric spectrum (<a href=\"https:\/\/www.slideshare.net\/seldon_io\/tensorflow-london-17-how-nasa-frontier-development-lab-scientists-use-ai-to-find-evidence-of-life\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>Then, it is time to retrieve the atmospheric parameters like temperature, pressure profile, atmospheric density, composing elements, etc. from spectrum. These are the biohints suggesting the planet is fit for life. For instance, the atmospheres\u2019 temperature provides an indication of the temperature at the surface. So, if the planet has the right temperature, there may be liquid water, which is one of the major requirements for hosting human-like life.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-artificial-intelligence-retrieving-atmosphere-from-spectrum.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-artificial-intelligence-retrieving-atmosphere-from-spectrum.png\" alt=\"\" width=\"640\" class=\"aligncenter size-full wp-image-37284\" \/><\/a><small>Retrieving atmospheric properties from spectrum (<a href=\"https:\/\/www.slideshare.net\/seldon_io\/tensorflow-london-17-how-nasa-frontier-development-lab-scientists-use-ai-to-find-evidence-of-life\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>These steps are challenging by themselves, and extra trouble adds up as no real data is available, so scientists have to find means to generate it, as well as processing and analyzing the data is computer-intensive. So, what&#8217;s the next move then?<\/p>\n<p>&nbsp;<\/p>\n<h3><span class=\"ez-toc-section\" id=\"Planetary_Spectrum_Generator\"><\/span>Planetary Spectrum Generator<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<p>To address the lack of real data, scientists at NASA use machine learning to generate it. To be more specific, NASA has a tool of its own\u2014<a href=\"https:\/\/psg.gsfc.nasa.gov\/help.php\" rel=\"noopener noreferrer\" target=\"_blank\">Planetary Spectrum Generator<\/a> (PSG)\u2014which provides a three-dimensional orbital calculator for most bodies in the Solar system and all the confirmed exoplanets. PSG is capable of calculating any possible geometry parameters needed when computing spectroscopic fluxes. The astronomical data is based on pre-computed ephemerides tables that provide orbital information from 1950 to 2050 with a precision of a minute.<\/p>\n<p>The PSG tool performs the numerical integration of the orbit by extracting orbital parameters from the <a href=\"https:\/\/exoplanetarchive.ipac.caltech.edu\/\" rel=\"noopener noreferrer\" target=\"_blank\">NASA Exoplanet Archive<\/a>. Due to the uncertainty and degeneracies in the derivation of the orbital parameters for exoplanets, PSG assumes the following:<\/p>\n<ul>\n<li style=\"margin-bottom: 6px;\">The longitude of ascending node (\u03a9) is assumed to be \u03c0.<\/li>\n<li style=\"margin-bottom: 6px;\">The planets are tidally locked, and the star sub-solar latitude\/longitude are set to the center of the planet.<\/li>\n<li>The phase identifies the true anomaly with respect to that of the secondary transit, with a phase of 180 degrees corresponding to the primary transit.<\/li>\n<\/ul>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-artificial-intelligence-generating-data-with-planetary-spectrum-generator.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-artificial-intelligence-generating-data-with-planetary-spectrum-generator.png\" alt=\"\" width=\"640\"  class=\"aligncenter size-full wp-image-37297\" \/><\/a><small>NASA uses its PSG tool to generate data (<a href=\"https:\/\/www.slideshare.net\/seldon_io\/tensorflow-london-17-how-nasa-frontier-development-lab-scientists-use-ai-to-find-evidence-of-life\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>To get a working model to be further used for training, scientists also apply <a href=\"https:\/\/www.altoros.com\/blog\/using-logistic-and-softmax-regression-with-tensorflow\/\" rel=\"noopener noreferrer\" target=\"_blank\">linear regression<\/a>, as well as <a href=\"https:\/\/en.wikipedia.org\/wiki\/Feedforward_neural_network\" rel=\"noopener noreferrer\" target=\"_blank\">feedforward<\/a> and <a href=\"https:\/\/www.altoros.com\/blog\/using-convolutional-neural-networks-and-tensorflow-for-image-classification\/\" rel=\"noopener noreferrer\" target=\"_blank\">convolutional neural networks<\/a>. After that, model grid search, selection, and tuning are conducted before the model is ready for training. William shed some light on the grid search parameters set, which are the following:<\/p>\n<ul>\n<li>learning rates: 0.0001, 0.001, 0.01<\/li>\n<li>optimizers: ADAM, SGD, ADAdelta, RMSProp<\/li>\n<li>activation functions: Tanh, Softmax, ReLU, ELU, Linear<\/li>\n<\/ul>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-Planetary-Spectrum-Generator-parameters.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-Planetary-Spectrum-Generator-parameters.png\" alt=\"\" width=\"640\" class=\"aligncenter size-full wp-image-37349\" \/><\/a><small>The parameters supported by the PSG tool (<a href=\"https:\/\/www.slideshare.net\/seldon_io\/tensorflow-london-17-how-nasa-frontier-development-lab-scientists-use-ai-to-find-evidence-of-life\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>So far, the team at NASA was able to generate around 100,000 data points for for training, 10,000 for validation, and around 7,710 for testing sets. This amounts to 2.5 million spectra at the training stage, 400,000 for validation, and 200,000 for testing.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-machine-learning-model-spectra.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-machine-learning-model-spectra.png\" alt=\"\" width=\"640\" class=\"aligncenter size-full wp-image-37306\" \/><\/a><small>Generating spectra data (<a href=\"https:\/\/www.slideshare.net\/seldon_io\/tensorflow-london-17-how-nasa-frontier-development-lab-scientists-use-ai-to-find-evidence-of-life\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>As William explained, in the exemplary results below, one can see atmospheric properties(on the right)\u2014how much water and methane is there\u2014generated for a single planet. The black lines indicate the prediction made through training a model, while the red lines indicate the real situation. As far as one can see, the results are quite matching. The closeups on the left give a more detailed insight into prediction vs. reality.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-machine-learning-to-generate-spectrum-data-results.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-machine-learning-to-generate-spectrum-data-results.png\" alt=\"\" width=\"640\" class=\"aligncenter size-full wp-image-37312\" \/><\/a><small>The results of generating spectrum data (<a href=\"https:\/\/www.slideshare.net\/seldon_io\/tensorflow-london-17-how-nasa-frontier-development-lab-scientists-use-ai-to-find-evidence-of-life\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>In <a href=\"https:\/\/gitlab.com\/frontierdevelopmentlab\/astrobiology\/pypsg\" rel=\"noopener noreferrer\" target=\"_blank\">this GitHub repo<\/a>, you will find a Python package for interacting with the Planetary Spectrum Generator.<\/p>\n<p>&nbsp;<\/p>\n<h3><span class=\"ez-toc-section\" id=\"INARA_Intelligent_Exoplanet_Atmosphere_Retrieval\"><\/span>INARA: Intelligent Exoplanet Atmosphere Retrieval<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<p>William also shared some information about another tool employed by NASA researchers\u2014<a href=\"https:\/\/gitlab.com\/frontierdevelopmentlab\/astrobiology\/inara\" rel=\"noopener noreferrer\" target=\"_blank\">INARA<\/a>, which stands for intelligent exoplanet atmosphere retrieval. This tool is used to generate spectra of rocky planets and to train a machine learning model for retrieving atmospheric parameters.<\/p>\n<p>INARA produces high resolution spectra and then saves observation simulation for the requested number of randomly generated planets across the parameters set. This mode allows for saving data for learning, validating, and testing a model. The generated spectra (observation simulations) can be used to train a machine learning model as a single one or in an ensemble mode.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-artificial-intelligence-to-find-exoplanets.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-artificial-intelligence-to-find-exoplanets.png\" alt=\"\" width=\"640\" class=\"aligncenter size-full wp-image-37343\" \/><\/a><small>Finding biohints to confirm an exoplanet as the one hosting life (<a href=\"https:\/\/www.slideshare.net\/seldon_io\/tensorflow-london-17-how-nasa-frontier-development-lab-scientists-use-ai-to-find-evidence-of-life\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>There is also a possibility to instantiate virtual machines (VMs) to either generate planetary spectra or use the above-mentioned PSG tool on the Google Cloud. The VMs are instantiated on hard-coded parameters loading Docker images and running the command-line parameters. You can check out how to install the tool and use it for different scenarios through <a href=\"https:\/\/gitlab.com\/frontierdevelopmentlab\/astrobiology\/inara\" rel=\"noopener noreferrer\" target=\"_blank\">its GitHub repo<\/a>.<\/p>\n<p>You can also make use of this <a href=\"https:\/\/calameo.com\/read\/0055032800dc29f522f5b\" rel=\"noopener noreferrer\" target=\"_blank\">cheat sheet<\/a>, which provides a quick introduction to classifying exoplanet candidates with machine learning.<\/p>\n<p>&nbsp;<\/p>\n<h3><span class=\"ez-toc-section\" id=\"Want_details_Watch_the_video\"><\/span>Want details? Watch the video!<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<table width=\"100%\">\n<tbody>\n<tr>\n<td class=\"video-details-td\">\n<div style=\"float:right; width:50%; padding-left:15px; font-size:14px;\">\n                <strong>Table of contents<\/strong><\/p>\n<ol>\n<li style=\"margin-bottom: 12px;\">The challenges of detecting exoplanets (02:41)<\/li>\n<li style=\"margin-bottom: 12px;\">Using the transit techniques to detect exoplanets (03:21)<\/li>\n<li style=\"margin-bottom: 12px;\">Measuring atmospheric spectrum (04:51)<\/li>\n<li style=\"margin-bottom: 12px;\">Retrieving atmosphere from the spectrum (05:52)<\/li>\n<li style=\"margin-bottom: 12px;\">Using the PSG tool to generate data (08:06)<\/li>\n<li style=\"margin-bottom: 12px;\">The exemplary machine learning model (10:21)<\/li>\n<li style=\"margin-bottom: 12px;\">Some results achieved (11:05)<\/li>\n<li style=\"margin-bottom: 12px;\">The INARA tool (13:51)<\/li>\n<li style=\"margin-bottom: 12px;\">Q&#038;As (14:15)<\/li>\n<\/ol><\/div>\n<div class=\"video-container\"><iframe loading=\"lazy\" title=\"TensorFlow London: How NASA scientists use AI to find evidence of life\" width=\"1200\" height=\"675\" src=\"https:\/\/www.youtube.com\/embed\/d6_N-sWXuwI?feature=oembed\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share\" referrerpolicy=\"strict-origin-when-cross-origin\" allowfullscreen><\/iframe><\/div>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p>&nbsp;<\/p>\n<h3><span class=\"ez-toc-section\" id=\"Related_slides\"><\/span>Related slides<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<p><center><iframe loading=\"lazy\" src=\"\/\/www.slideshare.net\/slideshow\/embed_code\/key\/I3zj5bEPVV6nnf\" width=\"595\" height=\"485\" frameborder=\"0\" marginwidth=\"0\" marginheight=\"0\" scrolling=\"no\" style=\"border:1px solid #CCC; border-width:1px; margin-bottom:5px; max-width: 100%;\" allowfullscreen> <\/iframe><\/div>\n<p><\/center><\/p>\n<p>&nbsp;<\/p>\n<h3><span class=\"ez-toc-section\" id=\"Further_reading\"><\/span>Further reading<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<ul>\n<li><a href=\"https:\/\/www.altoros.com\/blog\/analyzing-satellite-imagery-with-tensorflow-to-automate-insurance-underwriting\/\">Analyzing Satellite Imagery with TensorFlow to Automate Insurance Underwriting<\/a><\/li>\n<li><a href=\"https:\/\/www.altoros.com\/blog\/digital-twins-for-aerospace-better-fleet-reliability-and-performance\/\">Digital Twins for Aerospace: Better Fleet Reliability and Performance<\/a><\/li>\n<li><a href=\"https:\/\/www.altoros.com\/blog\/what-is-behind-deep-reinforcement-learning-and-transfer-learning-with-tensorflow\/\">What Is Behind Deep Reinforcement Learning and Transfer Learning with TensorFlow?<\/a><\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<h3><span class=\"ez-toc-section\" id=\"About_the_expert\"><\/span>About the expert<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<div>\n<div style=\"float: right;\"><a href=\"https:\/\/www.linkedin.com\/in\/william-fawcett\/\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/William-Fawcett-bio.png\" alt=\"\" width=\"150\" height=\"150\" class=\"aligncenter size-full wp-image-37161\" \/><\/a><\/div>\n<div style=\"width: 600px;\"><small><a href=\"https:\/\/www.linkedin.com\/in\/william-fawcett\/\" rel=\"noopener noreferrer\" target=\"_blank\">William Fawcett<\/a> is a researcher with a PhD in Particle Physics from the University of Oxford. He is skilled in data science, mathematical modeling, machine learning, and computer programming. William was engaged as a researcher in the NASA Frontier Development Lab, which is an 8-week applied artificial intelligence research accelerator established to maximize new AI technologies and capacities emerging in academia and in the private sector\u2014to apply them to challenges in the space sciences. William also serves as a research team leader at CERN, where he is involved in the analysis of LHC data for new physics. In his research, William makes use of data reduction techniques to deal with large data volumes (100 TB).<\/small><\/div>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<p>Houston, I want to believe!<\/p>\n<p>Being just a tiny blue dot in the ever-expanding universe, humans of Earth have been wondering for centuries if they are alone here. The answer to this question is the motivator behind scientific research of the world&#8217;s best minds, exquisitely equipped observatories, neatly designed spacecrafts, and [&#8230;]<\/p>\n","protected":false},"author":3,"featured_media":46833,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"content-type":"","footnotes":"","_links_to":"","_links_to_target":""},"categories":[26],"tags":[748],"class_list":["post-37154","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-use-cases","tag-machine-learning"],"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v26.6 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>How NASA Uses Artificial Intelligence to Detect Exoplanets | Altoros<\/title>\n<meta name=\"description\" content=\"There can be 40 billion stars potentially hosting human-like life. Learn how NASA uses transit techniques and machine learning to detect biohints on such stars.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"How NASA Uses Artificial Intelligence to Detect Exoplanets | Altoros\" \/>\n<meta property=\"og:description\" content=\"Houston, I want to believe! Being just a tiny blue dot in the ever-expanding universe, humans of Earth have been wondering for centuries if they are alone here. The answer to this question is the motivator behind scientific research of the world&#8217;s best minds, exquisitely equipped observatories, neatly designed spacecrafts, and [...]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/\" \/>\n<meta property=\"og:site_name\" content=\"Altoros\" \/>\n<meta property=\"article:published_time\" content=\"2018-10-10T19:29:24+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2019-09-20T17:48:11+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-artificial-intelligence-to-detect-exoplantes-machine-learning1.gif\" \/>\n\t<meta property=\"og:image:width\" content=\"640\" \/>\n\t<meta property=\"og:image:height\" content=\"360\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/gif\" \/>\n<meta name=\"author\" content=\"Sophia Turol\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Sophia Turol\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"8 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/\",\"url\":\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/\",\"name\":\"How NASA Uses Artificial Intelligence to Detect Exoplanets | Altoros\",\"isPartOf\":{\"@id\":\"https:\/\/www.altoros.com\/blog\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-artificial-intelligence-to-detect-exoplantes-machine-learning1.gif\",\"datePublished\":\"2018-10-10T19:29:24+00:00\",\"dateModified\":\"2019-09-20T17:48:11+00:00\",\"author\":{\"@id\":\"https:\/\/www.altoros.com\/blog\/#\/schema\/person\/58194952af19fe7b2b830846e077a58e\"},\"breadcrumb\":{\"@id\":\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#primaryimage\",\"url\":\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-artificial-intelligence-to-detect-exoplantes-machine-learning1.gif\",\"contentUrl\":\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-artificial-intelligence-to-detect-exoplantes-machine-learning1.gif\",\"width\":640,\"height\":360},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/www.altoros.com\/blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"How NASA Uses Artificial Intelligence to Detect Exoplanets\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.altoros.com\/blog\/#website\",\"url\":\"https:\/\/www.altoros.com\/blog\/\",\"name\":\"Altoros\",\"description\":\"Insight\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.altoros.com\/blog\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.altoros.com\/blog\/#\/schema\/person\/58194952af19fe7b2b830846e077a58e\",\"name\":\"Sophia Turol\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.altoros.com\/blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2019\/05\/trello_card-96x96.jpg\",\"contentUrl\":\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2019\/05\/trello_card-96x96.jpg\",\"caption\":\"Sophia Turol\"},\"description\":\"Sophia Turol is passionate about delivering well-structured articles that cater for picky technical audience. With 3+ years in technical writing and 5+ years in editorship, she enjoys collaboration with developers to create insightful, yet intelligible technical tutorials, overviews, and case studies. Sophie is enthusiastic about deep learning solutions\u2014TensorFlow in particular\u2014and PaaS systems, such as Cloud Foundry.\",\"url\":\"https:\/\/www.altoros.com\/blog\/author\/sophie-turol\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"How NASA Uses Artificial Intelligence to Detect Exoplanets | Altoros","description":"There can be 40 billion stars potentially hosting human-like life. Learn how NASA uses transit techniques and machine learning to detect biohints on such stars.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/","og_locale":"en_US","og_type":"article","og_title":"How NASA Uses Artificial Intelligence to Detect Exoplanets | Altoros","og_description":"Houston, I want to believe! Being just a tiny blue dot in the ever-expanding universe, humans of Earth have been wondering for centuries if they are alone here. The answer to this question is the motivator behind scientific research of the world&#8217;s best minds, exquisitely equipped observatories, neatly designed spacecrafts, and [...]","og_url":"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/","og_site_name":"Altoros","article_published_time":"2018-10-10T19:29:24+00:00","article_modified_time":"2019-09-20T17:48:11+00:00","og_image":[{"width":640,"height":360,"url":"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-artificial-intelligence-to-detect-exoplantes-machine-learning1.gif","type":"image\/gif"}],"author":"Sophia Turol","twitter_misc":{"Written by":"Sophia Turol","Est. reading time":"8 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/","url":"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/","name":"How NASA Uses Artificial Intelligence to Detect Exoplanets | Altoros","isPartOf":{"@id":"https:\/\/www.altoros.com\/blog\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#primaryimage"},"image":{"@id":"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#primaryimage"},"thumbnailUrl":"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-artificial-intelligence-to-detect-exoplantes-machine-learning1.gif","datePublished":"2018-10-10T19:29:24+00:00","dateModified":"2019-09-20T17:48:11+00:00","author":{"@id":"https:\/\/www.altoros.com\/blog\/#\/schema\/person\/58194952af19fe7b2b830846e077a58e"},"breadcrumb":{"@id":"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#primaryimage","url":"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-artificial-intelligence-to-detect-exoplantes-machine-learning1.gif","contentUrl":"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2018\/10\/NASA-using-artificial-intelligence-to-detect-exoplantes-machine-learning1.gif","width":640,"height":360},{"@type":"BreadcrumbList","@id":"https:\/\/www.altoros.com\/blog\/how-nasa-uses-artificial-intelligence-to-detect-exoplanets\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.altoros.com\/blog\/"},{"@type":"ListItem","position":2,"name":"How NASA Uses Artificial Intelligence to Detect Exoplanets"}]},{"@type":"WebSite","@id":"https:\/\/www.altoros.com\/blog\/#website","url":"https:\/\/www.altoros.com\/blog\/","name":"Altoros","description":"Insight","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.altoros.com\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Person","@id":"https:\/\/www.altoros.com\/blog\/#\/schema\/person\/58194952af19fe7b2b830846e077a58e","name":"Sophia Turol","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.altoros.com\/blog\/#\/schema\/person\/image\/","url":"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2019\/05\/trello_card-96x96.jpg","contentUrl":"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2019\/05\/trello_card-96x96.jpg","caption":"Sophia Turol"},"description":"Sophia Turol is passionate about delivering well-structured articles that cater for picky technical audience. With 3+ years in technical writing and 5+ years in editorship, she enjoys collaboration with developers to create insightful, yet intelligible technical tutorials, overviews, and case studies. Sophie is enthusiastic about deep learning solutions\u2014TensorFlow in particular\u2014and PaaS systems, such as Cloud Foundry.","url":"https:\/\/www.altoros.com\/blog\/author\/sophie-turol\/"}]}},"_links":{"self":[{"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/posts\/37154","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/comments?post=37154"}],"version-history":[{"count":57,"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/posts\/37154\/revisions"}],"predecessor-version":[{"id":46835,"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/posts\/37154\/revisions\/46835"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/media\/46833"}],"wp:attachment":[{"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/media?parent=37154"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/categories?post=37154"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/tags?post=37154"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}