This file contains invisible Unicode characters that may be processed differently from what appears below. If your use case is intentional and legitimate, you can safely ignore this warning. Use the Escape button to reveal hidden characters.
This file contains ambiguous Unicode characters that may be confused with others in your current locale. If your use case is intentional and legitimate, you can safely ignore this warning. Use the Escape button to highlight these characters.
---
title: "OpenPOWER Summit North America 2019: Counting the Stars - Neural Networks for Star Classification"
date: "2019-09-19"
categories:
- "blogs"
tags:
- "openpower-summit"
- "openpower-foundation"
- "openpower-summit-north-america"
- "atos"
- "counting-the-stars"
- "gaia-satellite"
- "university-of-geneva"
---
By: Hugh Blemings, Executive Director, OpenPOWER Foundation
The European Space Agency launched the [Gaia Satellite](https://sci.esa.int/web/gaia) in 2013. The satellite contains a 900 million-pixel camera and it takes a photograph of about 2 million stars every hour. As [Atos’](https://atos.net/en/) Jez Wain says, “it’s not your grand-dad’s digital camera.”
Attendees at [OpenPOWER Summit North America](https://events.linuxfoundation.org/events/openpower-summit-north-america-2019/) this year were treated to a session by Wain on how machine learning can be used to help classify stars. After all, there are 200 billion stars in the Milky Way, and the European Space Agency's Gaia project has only mapped about 1% of them. The stars now need to be classified in much the same way as we classify animals and plants. Atos is working with the University of Geneva to investigate the use of machine learning to help with this classification problem. Wain’s talk described the Gaia project before presenting the approach taken to construct and optimize a neural network capable of classifying the different star types.