Astronomy

When I heard Stephen Hawking extol the mysteries of black holes, I knew theoretical physics was what I wanted to do. There is still so much to learn about these strange regions, says Chanda Prescod-Weinstein

Feedback wonders if previous research into 'courtship behaviours of ostriches' in the UK will be taken into account by the owner of a new ostrich farm in New Hampshire

Ponds have long been neglected by science, but we can't overlook these diverse and important nature hotspots any more, say Jeremy Biggs and Penny Williams

To understand the power – and limitations – of artificial intelligence, we need information, not hype. Alex Wilkins explores what four new books, from Ray Kurzweil, Nick Bostrom, Neil Lawrence and Shannon Vallor, offer

The El Niño climate pattern has contributed to a year of record-breaking temperatures. We must bend the curve of carbon emissions before the next one arrives

The El Niño climate pattern has contributed to a year of record-breaking temperatures. We must bend the curve of carbon emissions before the next one arrives

We’ll soon be playing 'Star Wars: Hunters' on Nintendo Switch and mobile phones, and the latest gameplay trailer has a fun look at game modes and more.

"I say that to my team, whenever I have an opportunity. I share with my team that they are enabling science and exploration for dozens of missions being supported by NSN. Initially it just seems like words, but once they start realizing [their contributions] are real, I can tell you those people don't want to go anywhere. They just feel that sense of accomplishment." —Vir Thanvi, Deputy Program Manager, Exploration and Space Communications Projects Division, NASA's Goddard Space Flight Center

Sunspot region AR3697 has returned — unleashing yet another X-class solar flare. Watch it erupt from the sun's eastern limb here.

Cow udders have lots of bird-like flu virus receptors but no human-like ones, a study has found, meaning there’s no reason for the virus to evolve to become better at infecting people

Cow udders have lots of bird-like flu virus receptors but no human-like ones, a study has found, meaning there’s no reason for the virus to evolve to become better at infecting people

NASA astronauts Butch Wilmore and Suni Williams are back near their launch site in Florida ahead of their Crew Flight Test launch on June 1. A key flight readiness review today will confirm if Boeing Starliner is ready to go, after a helium leak.

Consumer-grade AI is finding its way into people’s daily lives with its ability to generate text and images and automate tasks. But astronomers need much more powerful, specialized AI. The vast amounts of observational data generated by modern telescopes and observatories defies astronomers’ efforts to extract all of its meaning.

A team of scientists is developing a new AI for astronomical data called AstroPT. They’ve presented it in a new paper titled “AstroPT: Scaling Large Observation Models for Astronomy.” The paper is available at arxiv.org, and the lead author is Michael J. Smith, a data scientist and astronomer from Aspia Space.

Astronomers are facing a growing deluge of data, which will expand enormously when the Vera Rubin Observatory (VRO) comes online in 2025. The VRO has the world’s largest camera, and each of its images could fill 1500 large-screen TVs. During its ten-year mission, the VRO will generate about 0.5 exabytes of data, which is about 50,000 times more data than is contained in the USA’s Library of Congress.

The VRO’s need for multiple sites to handle all of its data is a testament to the enormous volume of data it will generate. Without effective AI, that data will be stuck in a bottleneck. Image Credit: NOIRLab.

Other telescopes with enormous mirrors are also approaching first light. The Giant Magellan Telescope, the Thirty Meter Telescope, and the European Extremely Large Telescope combined will generate an overwhelming amount of data.

Having data that can’t be processed is the same as not having the data at all. It’s basically inert and has no meaning until it’s processed somehow. “When you have too much data, and you don’t have the technology to process it, it’s like having no data,” said Cecilia Garraffo, a computational astrophysicist at the Harvard-Smithsonian Center for Astrophysics.

This is where AstroPT comes in.

AstroPT stands for Astro Pretrained Transformer, where a transformer is a particular type of AI. Transformers can change or transform an input sequence into an output sequence. AI needs to be trained, and AstroPT has been trained on 8.6 million 512 x 512-pixel images from the DESI Legacy Survey Data Release 8. DESI is the Dark Energy Spectroscopic Instrument. DESI studies the effect of Dark Energy by capturing the optical spectra from tens of millions of galaxies and quasars.

AstroPT and similar AI deal with ‘tokens.’ Tokens are visual elements in a larger image that contain meaning. By breaking images down into tokens, an AI can understand the larger meaning of an image. AstroPT can transform individual tokens into coherent output.

AstroPT has been trained on visual tokens. The idea is to teach the AI to predict the next token. The more thoroughly it’s been trained to do that, the better it will perform.

“We demonstrated that simple generative autoregressive models can learn scientifically useful information when pre-trained on the surrogate task of predicting the next 16 × 16 pixel patch in a sequence of galaxy image patches,” the authors write. In this scheme, each image patch is a token.

This image illustrates how the authors trained AstroPT to predict the next token in a ‘spiralised’ sequence of galaxy image patches. It shows the token feed order. “As the galaxies are in the centre of each postage stamp, this set up allows us to seamlessly pretrain and run inference on differently sized galaxy postage stamps,” the authors explain. Image Credit: Smith et al. 2024.

One of the obstacles to training AI like AstroPT concerns what AI scientists call the ‘token crisis.’ To be effective, AI needs to be trained on a large number of quality tokens. In a 2023 paper, a separate team of researchers explained that a lack of tokens can limit the effectiveness of some AI, such as LLMs or Large Language Models. “State-of-the-art LLMs require vast amounts of internet-scale text data for pre-training,” the wrote. “Unfortunately, … the growth rate of high-quality text data on the internet is much
slower than the growth rate of data required by LLMs.”

AstroPT faces the same problem: a dearth of quality tokens to train on. Like other AI, it uses LOMs or Large Observation Models. The team says their results so far suggest that AstroPT can solve the token crisis by using data from observations. “This is a promising result that suggests that data taken from the observational sciences would complement data from other domains when used to pre-train a single multimodal LOM, and so points towards the use of observational data as one solution to the ‘token crisis’.”

AI developers are eager to find solutions to the token crisis and other AI challenges.

Without better AI, a data processing bottleneck will prevent astronomers and astrophysicists from making discoveries from the vast quantities of data that will soon arrive. Can AstroPT help?

The authors are hoping that it can, but it needs much more development. They say they’re open to collaborating with others to strengthen AstroPT. To aid that, they followed “current leading community models” as closely as possible. They call it an “open to all project.”

“We took these decisions in the belief that collaborative community development paves the fastest route towards realising an open source web-scale large observation model,” they write.

“We warmly invite potential collaborators to join us,” they conclude.

It’ll be interesting to see how AI developers will keep up with the vast amount of astronomical data coming our way.

The post Astronomy Generates Mountains of Data. That’s Perfect for AI appeared first on Universe Today.

Just like the fictional planet of Vulcan was wiped out in Star Trek, new research has destroyed the real-life version of Spock's homeworld, albeit in a less violent fashion.

People who have had a recent vaccine against tetanus appear to be less likely to develop Parkinson’s disease, suggesting that the bacterial infection is involved in the condition

People who have had a recent vaccine against tetanus appear to be less likely to develop Parkinson’s disease, suggesting that the bacterial infection is involved in the condition

Warming waters are reaching several miles into Antarctica’s Thwaites Glacier—nicknamed the “doomsday glacier” because of its potential impact on sea-level rise

Video: 00:03:29

Mars’s surface is covered in all manner of scratches and scars. Its many marks include the fingernail scratches of Tantalus Fossae, the colossal canyon system of Valles Marineris, the oddly orderly ridges of Angustus Labyrinthus, and the fascinating features captured in today’s video release from Mars Express: the cat scratches of Nili Fossae.

Nili Fossae comprises parallel trenches hundreds of metres deep and several hundred kilometres long, stretching out along the eastern edge of a massive impact crater named Isidis Planitia.

This new video features observations from Mars Express's High Resolution Stereo Camera (HRSC). It first flies northwards towards and around these large trenches, showing their fractured, uneven appearance, before turning back to head southwards. It ends by zooming out to a ‘bird’s eye’ view, with the landing site of NASA’s Perseverance rover, Jezero Crater, visible in the lower-middle part of the final scene. (You can explore this crater further via ESA’s interactive map.)

The trenches of Nili Fossae are actually features known as ‘graben’, which form when the ground sitting between two parallel faults fractures and falls away. As the graben seem to curve around Isidis Planitia, it’s likely that they formed as Mars’s crust settled following the formation of the crater by an incoming space rock hitting the surface. Similar ruptures – the counterpart to Nili Fossae – are found on the other side of the crater, and named Amenthes Fossae.

Scientists have focused on Nili Fossae in recent years due to the impressive amount and diversity of minerals found in this area, including silicates, carbonates, and clays (many of which were discovered by Mars Express’s OMEGA instrument). These minerals form in the presence of water, indicating that this region was very wet in ancient martian history. Much of the ground here formed over 3.5 billion years ago, when surface water was abundant across Mars. Scientists believe that water flowed not only across the surface here but also beneath it, forming underground hydrothermal flows that were heated by ancient volcanoes.

Because of what it could tell us about Mars’s ancient and water-rich past, Nili Fossae was considered as a possible landing site for NASA’s Curiosity rover, before the rover was ultimately sent to Gale Crater in 2012. Another mission, NASA’s Perseverance rover, was later sent to land in the nearby Jezero Crater, visible at the end of this video.

Mars Express has visited Nili Fossae before, imaging the region’s graben system back in 2014. The mission has orbited the Red Planet since 2003, imaging Mars’s surface, mapping its minerals, studying its tenuous atmosphere, probing beneath its crust, and exploring how various phenomena interact in the martian environment. For more from the orbiter and its HRSC, see ESA's Mars Express releases.

Disclaimer: This video is not representative of how Mars Express flies over the surface of Mars. See processing notes below.

Processing notes: The video is centred at 23°N, 78°E. It was created using Mars Chart (HMC30) data, an image mosaic made from single-orbit observations from Mars Express’s HRSC. This mosaic was combined with topography derived from a digital terrain model of Mars to generate a three-dimensional landscape. For every second of the movie, 62.5 separate frames are rendered following a pre-defined camera path. The vertical exaggeration is three-fold. Atmospheric effects – clouds and haze – have been added, and start building up at a distance of 50 km.

Click here for the original video created by Freie Universität Berlin, who use Mars Express data to prepare spectacular views of the martian surface. The original version has no voiceover, captions or ESA logo.

As ESA’s Hera spacecraft for planetary defence goes through pre-flight testing, the system that will steer it around its target binary asteroid system is also undergoing its final checks for space.

Wearing earplugs, hearing aids and earphones can make your own voice sound booming, but a new design dampens the din