Scientists taught artificial intelligence to predict the distribution of dark matter.
A team of astrophysicists from the United States and South Korea has used artificial intelligence to create a new map of the distribution of dark matter. The neural network has revealed mysterious thread-like structures that link galaxies like a spider web.
10 things you should know about the dark matter map and bridges between galaxies
1. The generally accepted theory is that dark matter accounts for 80 percent of all matter in the universe. It is invisible to the human eye, and even with the help of existing equipment, it has not yet been possible to document it.
2. Some theorize that it may consist of neutral particles that may not interact with anything on the electromagnetic spectrum, as for example, light, which is why we do not see it.
3. However, it is believed that it is dark matter that constitutes the skeleton of a large-scale structure called the cosmic web. The cosmic web is what dictates the movement of matter in space and understanding the distribution of dark matter is perhaps the key piece to solving countless cosmic mysteries.
4. In their research, astrophysicists took a fundamentally new approach using machine learning. They trained the neural network to build a model based on the available data on the distribution and movement of galaxies.
5. Thanks to this, artificial intelligence has learned to predict the distribution of dark matter. Doing this with something unproven and invisible is difficult and we can consider it speculative. Yet, it is curious to see and get a better idea of the distribution of dark matter on a cosmic map.
6. To train the neural network, the scientists used a large set of galaxy simulations called Illustris-TNG. Moreover, simulated galaxies were specially selected, comparable in their characteristics with our Milky Way.
7. This helped determine exactly what galactic properties are needed to predict the distribution of dark matter.
8. With help of data on the motion of galaxies and their distribution in space, the neural network was able to improve the quality and reveal additional details on the dark matter map.
9. The researchers then applied their model to real data from the Cosmicflow-3 catalog of galaxies. As a result, both known visible structures in the local Universe and previously unknown structures were consistently displayed on the map.
10. In particular, artificial intelligence has displayed on the map relatively small thread-like structures connecting galaxies. The authors of the work called these structures bridges between galaxies. They, of course, now require further study.
Studying and understanding dark matter is essential to understand both the past and future of the universe. It dictates the movements of galaxies and any objects in space. Therefore, knowing the distribution of dark matter in an accurate map, for instance, would give us a clear look into the future of the universe and where our Milky Way will be located billions of years from now.
Unfortunately, no current technology has made it possible to see/find dark matter, which remains the main task ahead. There are scientists that still doubt its existence because it does not apply to certain written laws. Yet, all the necessary evidence apart from its clear detection exists. Perhaps the new generation of space observatories will finally solve this mystery in the upcoming years.
Join the discussion and participate in awesome giveaways in our mobile Telegram group. Join Curiosmos on Telegram Today. t.me/Curiosmos
• Futurity. (2021, May 27). Dark matter map shows hidden ‘bridges’ connect galaxies. Futurity.
• Hong, S. E., Jeong, D., Hwang, H. S., & Kim, J. (2021, March 18). Revealing the Local Cosmic Web from Galaxies by Deep Learning. arXiv.org.
• Pappas, S. (2021, May 28). Dark matter map reveals new filaments connecting galaxies. LiveScience.
• Stephanie Pappas, L. S. (n.d.). New Dark Matter Map Shows The Bridges Between The Milky Way And Nearby Galaxies. ScienceAlert.