Artificial intelligence to measure dark matter

Los cúmulos de galaxias are the most massive astronomical objects in the universe and are made up to 85% of dark matter (matter that does not emit light and whose nature we still do not know).

An international collaboration led by astrophysicists from the Universidad Autónoma de Madrid (UAM) in Spain has estimated, for the first time through artificial intelligence techniques, the total matter content of a large set of galaxy clusters.

The technique employed is of the type known as automatic learning (a modality of artificial intelligence).

An automatic learning algorithm based on convolutional neural networks, a revolutionary model developed for automatic image processing, has been used for the study.

The results of the use of the algorithm in this study include the quantities of dark matter and are based on observations in the microwave range of the Planck space mission of the ESA (European Space Agency).

“The procedure consists of compressing the information contained in the images, so that it can be easily compared with some internal property of the image, which in this case would be the total mass of the cloud that occupies the image”, explains Daniel de Andrés, assistant professor of the Department of Theoretical Physics of the UAM and prime firmante del trabajo.

To train the algorithm, the researchers had to generate almost 200,000 images equivalent to how the Planck satellite would observe a set of numerically simulated cumulus clouds.

These synthetic images were obtained from the results of the numerical simulations of the project The Three Hundred, led also by the researchers of the UAM.

“Once the convolutional network was trained, it was applied to more than a thousand real images of galaxy clusters observed by the Planck satellite, resulting in the estimation of the total mass of the astronomical object, according to what the algorithm learned de las simulaciones, pero en este caso, libres de posibles sesgos deboto a las hypotesis teóricas que se asumen en estudios previos”, explains Daniel de Andrés.

[Img #67244]

Artistic recreation of the concept of artificial intelligence used to estimate the content of dark and normal matter in clusters of galaxies. (Image: UAM)

Por su parte, the co-author Weiguang Cui, affirms that “los resultados obtenidos son muy excitantes. Artificial intelligence is a tool that will help us understand the complex relationships between the different components of matter that occupy the universe.”

In the collaboration, researchers from the University of Edinburgh, the Instituto Tecnológico de Massachusetts (MIT) and the University of Sapienza of Rome, together with experts from EURA NOVA, a company based in Belgium and specialized in automatic learning techniques, are participating.

Aprendizaje automático y astrofísica

Artificial intelligence, and in particular el aprendizaje automático, facilitates that a machine can make predictions from the study of large quantities of data. In this form, los ordenadores pueden realizar tasks complejas, como recognar objetos en imágenes o jugar al jedrez.

El aprendizaje automático has given place recently to multiple applications in different fields of industry and research. Not only is it useful to accelerate the analysis of enormous amounts of data, but it also brings new discoveries.

In particular, the methods of artificial intelligence for the analysis of astrophysical data are being used every time in the most habitual way in the processing of astronomical images.

Cúmulos de galaxias y materia oscura

Los cúmulos de galaxias son agrupaciones de centenares o miles de galaxias enlazados gravitationally. The total matter that generates this gravitational field is formed, in more than 85%, by a component of nature that we still do not understand, and that is called materia oscura.

Hasta ahora, la materia total de estos objetos se había medido principalum de dos maneras. A partir de observaciones de la emission en rayos X del gas intracumular. Y mediate las distortiones del espectro de frecuencias de los photones de la radiación de fundo cósmica, cuando atraviesan el gas caliente del cúmulo en su recorrido hacia nuestros radiotelescopios.

“In both cases, from the two-dimensional images that we receive with our instruments, we have to apply certain theoretical hypotheses about the state of the cumulus gas to be able to deduce the total mass that it contains,” explains Daniel de Andrés.

“The only form that we had until now to establish these theoretical relations between the emission of gas and the mass was based on numerical simulations that tried to reproduce the formation of these objects and that modeled the complex relations that exist between the various components of the cluster: gas, estrellas, supernovas, agujeros negros y materia oscura”, concluded the researcher. (Fuente: UAM)


Leave a Comment

Your email address will not be published. Required fields are marked *