News

“KI Lectures”: LMU astrophysicist sees AI as essential to exploring the universe

21 Jan 2022

In cosmological research, advances in telescopes and cameras are generating enormous amounts of data. Classical algorithms are not up to the task of processing this flood of data — new methods are needed

“Artificial intelligence is the key to new discoveries in cosmology,” said Professor Daniel Grün, chair of astrophysics, cosmology and artificial intelligence at LMU. Speaking in one of LMU’s “KI Lectures,” the astrophysicist highlighted the potential of artificial intelligence applications for exploring the universe.

In cosmology, enormous amounts of data are being collected through telescopes and the latest camera technology. “Volumes of astronomical data are growing exponentially,” says Daniel Grün. Computers are no longer powerful enough to process them like they did before; they are currently the limiting factor for astronomical research. Artificial intelligence algorithms, for their part, can be used as tools to evaluate the mountains of data.

Questioning the previous model of the universe with algorithms

News

The universe in the machine brain: Artificial Intelligence in Cosmology

Read more

“Within the next decade, advances in camera and telescope development will make it possible to take images of the entire night sky and observe almost all the galaxies that exist in the universe,” says Daniel Grün. What this also means is that better algorithms are needed to evaluate the data and analyze the structures of the universe. The statistical requirements are very diverse, he adds. “We can develop artificial intelligence that meets these exact requirements. That’s the key to uncovering new insights in cosmology.”

The astrophysicist is counting on the fact that as better algorithms are developed, it will also be possible to better verify the current physical laws of the universe. “We have a very successful model of our universe that accurately describes all the observations that have been made. But at the same time, we know that this model is wrong, that it is incomplete,” says Grün. “AI can help us test physical theories with greater accuracy.”

Explainable AI, which enables us to see how results have been obtained, also plays a crucial role here: “We need to understand what AI draws its conclusions from.”

The lecture and the subsequent discussion with Prof. Oliver Jahraus, LMU Vice President, and the audience is available on the LMU YouTube channel.

For further dates for the “KI Lectures” and to register, click here

For more information on Daniel Grün's research, see

What are you looking for?