Scientists use new deep learning method to add 301 planets to Kepler’s total number

Deep neural networks are machine learning systems that automatically learn a task if the necessary data is provided. An artificial neural network (ANN) with many layers between the input and output layers is called a deep neural network (DNN). Neural networks come in different shapes and sizes. However, they all include the same essential components: neurons, synapses, weights, biases, and functions.

Recently, scientists added a total of 301 validated exoplanets to the already existing count of exoplanets. The Planet Cluster is the most recent addition to the 4,569 confirmed planets orbiting various distant stars. This news made everyone wonder how scientists could discover such a huge all at once. The answer lies in a deep neural network called ExoMiner. The Kepler Science Office upgraded the 301 machine validated planets to candidate planet status after discovering them through the Kepler Science Operations Center pipeline. None of this was possible before the implementation of ExoMiner.

ExoMiner is a recent development in deep neural networks that uses NASA’s Pleiades supercomputer to detect true exoplanets from various forms of “false positives”. Its design is based on the tests and qualities that human scientists use to confirm the discovery of new exoplanets. It also learns from previously verified exoplanets as well as false positive cases.

Going through the data and deciphering what is not a real planet forms the basis of ExoMiner. There are thousands of stars in the field of view of projects like NASA’s Kepler and its tracking mission, K2, each with the potential to host many possible exoplanets. Looking at huge data sets is a time consuming process. ExoMiner, on the other hand, also takes care of that. Machine learning programs for detecting exoplanets, in general, are like black boxes; no one knows how they decide if something is a planet or not. However, scientists have provided specific insight into aspects of the data that led the neural network to validate or reject a planet when using ExoMiner.

ExoMiner discovered all 301 planets using data from the remaining set of potential – or candidate – planets in the Kepler Archives, according to a research article published in the Astrophysical Journal. The article also reveals how ExoMiner is more accurate and reliable at ruling out false positives and displaying actual signs of planets orbiting their mother stars. It does all of this while allowing scientists to understand precisely how ExoMiner came to its conclusions.

In addition to confirming 301 planets, the research team pointed out that none of them look like Earth. However, the newly discovered planets share similar characteristics to the global population of confirmed exoplanets in our galactic neighborhood.

THE RESEARCH FOR EXOPLANETS CONTINUES …

ExoMiner was also trained with Kepler data. The knowledge gained can be applied to other assignments including TESS, the team’s current effort.

ExoMiner will have a chance to prove its mettle when missions like NASA’s Transiting Exoplanet Survey Satellite, or TESS, and Planetary Transits and Oscillations of Stars, or PLATO, which use transit photometry, open the door to more. opportunities.

The references:

  • https://www.jpl.nasa.gov/news/new-deep-learning-method-adds-301-planets-to-keplers-total-count
  • https://phys.org/news/2021-11-deep-method-planets-kepler-total.html

Source link

About Donald P. Hooten

Check Also

The 369 Manifestations Method: How to Make Your Therapy Sessions Work

Psychologist Nancy Sokarno walks us through the 369 Manifestations Method and how to get the …