Here's How AI is Helping Astronomers to Understand Universe's Fundamental Parameters

Artificial intelligence (AI) is making waves beyond the realm of digital assistants and deepfake technology. Astronomers are now harnessing AI to achieve unprecedented precision in understanding the universe’s fundamental parameters. Researchers at the Flatiron Institute’s Center for Computational Astrophysics (CCA) in New York City have leveraged AI to calculate the five cosmological parameters that characterise the universe, enhancing our knowledge of its structure and evolution.

Unveiling the Universe’s Settings

The five cosmological parameters in question are crucial as they essentially define the ‘settings’ of the universe, dictating how it operates on the largest scales. Liam Parker, an astronomer at CCA and a co-author of the study, explained that these parameters are like the universe’s operating instructions. To determine these parameters with exceptional accuracy, the research team employed AI to analyse data from over 100,000 galaxies, which were observed as part of the Sloan Digital Sky Survey (SDSS). This survey is instrumental in cosmology as it helps map out the universe by examining how galaxies are distributed across space.

Maximising Data Efficiency

The use of AI allowed the team to extract detailed insights from the SDSS data, a task previously hampered by the sheer volume of information and the limitations of traditional analysis methods. Co-author Shirley Ho, another astronomer at CCA, highlighted the enormous costs associated with such surveys, which can run into billions of dollars. With AI, the team has been able to push the boundaries of what these surveys can reveal, providing a more cost-effective means of extracting valuable knowledge from complex data.

Training AI for Precision

The process involved training the AI model on 2,000 simulated universes, each with different cosmological settings. This training set included real-world challenges such as atmospheric distortion and imperfect telescope imaging to ensure the AI’s accuracy. Once trained, the AI was applied to actual data from the SDSS Baryon Oscillation Spectroscopic Survey. The results were impressive: the AI reduced the uncertainty in measuring the universe’s ‘clumpiness’ by more than half compared to traditional methods. This achievement is akin to conducting traditional analyses with four times more data, significantly enhancing data efficiency.

Impact on Cosmic Mysteries

The advancement of AI in cosmology is not just about refining existing methods but also about addressing major cosmic questions. One such mystery is the Hubble tension, which refers to discrepancies between different estimates of the Hubble constant, a measure of the universe’s expansion rate. The improved precision provided by AI could be crucial in resolving this tension and advancing our understanding of the universe.

As new surveys like the European Euclid survey come online, the AI-powered techniques developed by the CCA team will be instrumental in extracting maximum value from these new datasets. The ability to analyse and interpret vast amounts of cosmic data efficiently represents a significant leap forward in our quest to comprehend the universe’s fundamental parameters.