AI helps detect new space aberrations

outer space

Credit: Pixabay/CC0 Public Domain

The SNAD team, an international network of researchers including Matvey Kornilov, assistant professor in the School of Physics at HSE University, discovered 11 previously undiscovered space anomalies, seven of which are supernova candidates. The researchers analyzed digital images of the northern sky taken in 2018 using the kD tree to detect anomalies through the “nearest neighbour” method. Machine learning algorithms have helped to automate the search. The paper was published in new astronomy.

Most astronomical discoveries were based on observations with later calculations. While the total number of observations in the 20th century was still relatively small, data volumes increased dramatically with the arrival of large-scale astronomical surveys. For example, the Zwicky Transit Facility (ZTF), which uses a wide-field vision camera to survey the northern sky, generates 1.4 terabytes of data on observation night and its catalog contains billions of objects. Processing such huge amounts of data manually is expensive and time-consuming, so the SNAD team of researchers from Russia, France and the United States came together to develop an automated solution.

When scientists study astronomical bodies, they observe their light curves, which show variations in an object’s brightness as a function of time. Observers first recognize a flash of light in the sky and then follow its evolution to see if the light gets brighter or weaker over time, or turns off. In this study, researchers examined one million real light curves from the 2018 ZTF catalog and seven simulated live curve models for the species under study. In total, they followed about 40 parameters, including the amplitude of the brightness of the object and the time frame.

“We describe the characteristics of our simulations using a range of properties that would be expected to be observed in real astronomical objects. In the data set of nearly a million objects, we were looking for ultra-strong supernovae, Type I supernovae, and Type II supernovae. , and tides and turbulence events,” explains Konstantin Malanchev, co-author of the research paper and a postdoctoral researcher at the University of Illinois at Urbana-Champaign. “We refer to these classes of objects as anomalies. They are either very rare, with unknown properties, or they seem interesting enough to merit further study.”

Then the light curve data from the real objects were compared with the simulation data using the kD tree algorithm. A kD tree is a geometric data structure for dividing space into smaller parts by clipping them using hyperplanes, planes, lines, or points. In the current research, this algorithm was used to narrow the search when searching for real objects with characteristics similar to those shown in the seven simulations.

Next, the team identified 15 closest neighbors, meaning real objects from the ZTF database, for each simulation — a total of 105 matches, which the researchers then visually scanned for anomalies. Manual verification confirmed 11 anomalies, seven of which were supernova candidates, and four were candidates for active galactic cores where tidal perturbation events could occur.

“This is a very good result,” comments Maria Prozinskaya, a co-author on the paper and a research fellow at the Sternberg Astronomical Institute. “In addition to the rare objects already discovered, we were able to discover many new things that astronomers had previously lost. This means that existing search algorithms can be improved to avoid losing such objects.”

This study shows that the method is very effective, while it is relatively easy to apply. The proposed algorithm for detecting space phenomena of a certain type is universal and can be used to detect any astronomical objects of interest, and is not limited to rare types of supernovae.

“Astronomical phenomena and astrophysics that have not yet been discovered are actually anomalies,” said Matvey Kornilov, associate professor in the School of Physics at HSE University. “It is expected that their observed manifestations will differ from the properties of known objects. In the future, we will try to use our method to discover new classes of objects.”


New anomaly detection pipeline for astronomical detection and recommendation systems


more information:
PD Aleo et al, SNAD transient miner: finding the missing transient events in ZTF DR4 using kD trees, new astronomy (2022). DOI: 10.1016 / j.newast.2022.101846

Provided by the National Research University Higher School of Economics

the quote: AI Helps Detect New Space Aberrations (2022, Aug 5) Retrieved Aug 5, 2022 from https://phys.org/news/2022-08-ai-space-anomalies.html

This document is subject to copyright. Notwithstanding any fair dealing for the purpose of private study or research, no part may be reproduced without written permission. The content is provided for informational purposes only.

Leave a Comment