A successor to the Hubble Space Telescope, the James Webb Space Telescope is the largest and most powerful space observatory ever built. Image: ESA/ATG medialab
A machine learning model called Morpheus will be used to detect and classify galaxies in deep space, to help map the earliest structures in the universe.
The first images from the James Webb Space Telescope are set to be released on 12 July, and it is expected that AI will play a key role in analysing the received data.
A successor to the Hubble Space Telescope, the James Webb Space Telescope (JWST) is the largest and most powerful space observatory ever built.
Scientists are busy working on machine learning models to help analyse and quickly classify the vast amounts of data to come from their new eye on the cosmos.
“What’s great about AI and machine learning is that you can train a model to actually make those decisions for you in a way that is less hands-on and more based on a set of metrics that you define,” Prof Brant Robertson told Nvidia’s chief blogger Brian Caulfield.
Robertson, an astronomy and astrophysics professor at UC Santa Cruz, is one of the scientists behind Morpheus, a powerful AI for detecting and classifying galaxies.
Morpheus is a machine learning model that generates pixel-level morphological classifications of astronomical sources. Tech giant Nvidia said that its GPUs will help accelerate Morpheus across multiple platforms.
Morpheus was previously used to help classify images for the Hubble Space Telescope. With a growing team and advancements in machine learning, Robertson said the AI model will be used for “all of the major extragalactic JWST surveys” conducted in its first year.
The COSMOS-Webb project
Robertson and a team of nearly 50 researchers will be using the AI model as part of the COSMOS-Webb project. This is an ambitious plan to use JWST to map the earliest structures in the universe.
The telescope will survey a large patch of the sky with its near-infrared and mid-infrared cameras, across more than 200 hours of observing time.
As part of the project, the researchers plan to survey half a million galaxies with multiband, high-resolution, near-infrared imaging, along with 32,000 galaxies in mid-infrared imaging.
“The COSMOS-Webb project is the largest contiguous area survey that will be executed with JWST for the foreseeable future,” Robertson said.
It aims to build on earlier discoveries to advance our knowledge of the universe, such as learning how dark matter evolved over time and looking for early, fully evolved galaxies.
“The JWST will really enable us to see the universe in a new way that we’ve never seen before,” Robertson said. “So it’s really exciting.”
10 things you need to know direct to your inbox every weekday. Sign up for the Daily Brief, Silicon Republic’s digest of essential sci-tech news.
