A more precise comprehension of dark energy has been attained through the utilization of artificial intelligence (AI) techniques by a research team led by University College London. The team employed AI to analyze a map encompassing dark and visible matter in the universe over the past 7 billion years, enhancing the accuracy of key characteristics, such as the density of dark energy, by a factor of two. This advancement enabled the dismissal of previously plausible universe models.
Dark energy, constituting about 70% of the universe’s content, is the enigmatic force propelling the universe’s expansion, while dark matter and normal matter make up 25% and 5%, respectively. The study’s lead author, Dr. Niall Jeffrey, highlighted the significance of employing AI to extract insights from computer-simulated universes, emphasizing the efficiency gained compared to traditional methods that would necessitate significantly more data collection.
The research outcomes align with the prevailing notion of dark energy as a ‘cosmological constant,’ yet they also allow for alternative explanations, such as potential flaws in the theory of gravity. The analysis of the Dark Energy Survey map indicated a more uniform distribution of matter in the universe than predicted by Einstein’s general relativity, though with less pronounced deviations compared to prior studies due to larger error margins.
The Dark Energy Survey map, derived through weak gravitational lensing, involved observing the deflection of light from distant galaxies by intervening matter’s gravity. By scrutinizing distortions in the shapes of 100 million galaxies, the researchers inferred the distribution of all matter, both visible and dark, in the foreground galaxies, covering a significant portion of the Southern Hemisphere’s sky.
Utilizing UK government-funded supercomputers, the team conducted simulations based on the Dark Energy Survey data to generate various universe models. Machine learning algorithms were then employed to extract pertinent information from these simulations and evaluate the compatibility of observed data with different cosmological models. This innovative approach allowed for a more comprehensive analysis of the data than previously feasible.
The forthcoming dark universe initiatives, including the European Space Agency’s Euclid mission, are poised to provide additional data on the universe’s large-scale structures, aiding in the assessment of current cosmological models. These endeavors seek to elucidate whether the observed uniformity of the universe contradicts predictions based on the cosmic microwave background analysis, shedding light on the fundamental nature of dark energy and the universe’s evolution.
