Artificial intelligence has been trained to find new exploding stars, identify new types of galaxies, and detect the mergers of massive stars, which is faster than the rate of new discovery in the oldest science.
Astronomers at the University of California, Berkeley found hidden connections in the mathematics of general relativity, which can be used to find new planets around other stars.
In a paper appearing this week in the journal Nature Astronomy, the researchers describe how an artificial intelligence program developed to more quickly detect exoplanets when planetary systems pass in front of a background star.
In 1936, Albert Einstein used his new theory of general relativity to show how the light from a distant star can be bent by the gravity of a foreground star, and often splitting it into several points of light. A hand lens can focus and intensify light from the sun.
The light curve is more complicated when the object is a star with a planet. Degeneracies are when there are multiple planetary orbits that can explain a light curve equally well. Humans simplified the math and missed the bigger picture.
The researchers admit that the twoories are probably still incomplete, but they pointed to a mathematical way to unify the two major kinds of degeneracy in interpreting what telescopes detect during microlensing.
The paper was uploaded to a preprint server last year, after a machine learning inference algorithm led to the discovery of something new and fundamental about the equations that govern the general relativistic effect of light bending by two massive bodies. A professor of astronomy at UC Berkeley, Bloom is the chair of the department.
He compared the discovery by UC Berkeley graduate student to the connections that DeepMind made between two different areas of mathematics. These examples show that there are associations that humans miss.
It is argued that they are one of the first times that artificial intelligence has been used to yield new theoretical insight in math and astronomy.
Scott Gaudi, a professor of astronomy at The Ohio State University, is one of the pioneers of using gravitational microlensing to discover exoplanets. This is indicative of how research will go in the future when it is aided by machine learning, which is really exciting.
Microlensing is being used to discover exoplanets.
More than 5,000 exoplanets, or extrasolar planets, have been discovered around stars in the Milky Way, but few have actually been seen through a telescope. Most have been detected because they create a Doppler wobble in the motions of their host stars or because they slightly dim the light from the host star when they cross in front of it. Microlensing has discovered a few more than 100.
One of the main goals of NASA's Nancy Grace Roman Space Telescope is to discover thousands more exoplanets via microlensing. The technique has an advantage over the other techniques in that it can detect planets that are far away from their stars at a distance equivalent to that of Jupiter or Saturn in our solar system.
Two years ago, Bloom,Zhang and their colleagues set out to develop an artificial intelligence method to analyze microlensing data faster to determine the stellar and planetary mass of these planetary systems. The Roman telescope will detect hundreds of thousands of events in order to find the 1% or fewer that are caused by exoplanetary systems.
The observed signal can be ambiguous. When a lone foreground star passes in front of a background star, the brightness of the background stars rises to a peak and then drops symmetrically to its original brightness. It is easy to understand.
If the foreground star has a planet, it creates a separate peak within the peak caused by the star. When reconstructing the orbital configuration of the exoplanet that produced the signal, general relativity often allows two or more so-called degenerate solutions, all of which can explain the observations.
Astronomers have dealt with these degeneracies in simplistic ways. If the distant starlight passes close to the star, the observations could be interpreted as a wide or a close circle for the planet. When the background starlight passes close to the planet, there is a second type of degeneracy. The two different solutions for the planetary circle are not very different.
Two-body microlensing can be used to determine the true mass and orbital distances. In a paper published last year, a group of UC Berkeley co-authors, including astronomy professor Jessica Lu, described a new artificial intelligence that does not rely on knowledge of these interpretations at all. The analysis of microlensing observations can be done in less time and with less computer crunching.
There were other interpretations that the two interpretations did not account for, and that's what Zhang noticed when he tested the new artificial intelligence algorithm on microlensing light curves from hundreds of possible orbital configurations of star and exoplanet. He concluded that the commonly used interpretations of microlensing were just special cases of a broader theory that explains the full variety of ambiguities in microlensing events.
The two previous theories deal with cases where the background star appears to pass close to the foreground star or the foreground planet. The new unifying theory was based on that.
After seeing many examples where the previous two theories did not fit observations, Gaudi came around after seeing the new theory. The data from two dozen previous papers that reported the discovery of exoplanets through microlensing was looked at by Zhang and he found that the new theory fit the data better than the previous theories.
The people were seeing the microlensing events, but they didn't realize it.
The new mathematics based on general relativity and the theory of microlensing situations where more than one exoplanet is circling a star have been described in a new paper.
Interpretation of microlensing observations is made more ambiguous by the new theory. The theory shows that observing the same microlensing event from two perspectives, from Earth and from the Roman Space Telescope, will make it easier to settle on the correct orbits and mass. Astronomers currently plan to do that.
The artificial intelligence suggested a way to look at the lens equation in a new light and uncover something really deep about the mathematics of it. The two were able to do something with an expert like Keming.
More information: Keming Zhang et al, A ubiquitous unifying degeneracy in two-body microlensing systems, Nature Astronomy (2022). DOI: 10.1038/s41550-022-01671-6 Journal information: Nature Astronomy Citation: AI reveals unsuspected math underlying search for exoplanets (2022, May 24) retrieved 24 May 2022 from https://phys.org/news/2022-05-ai-reveals-unsuspected-math-underlying.html This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.
All Rights Reserved © 2024
