Artist’s impression: SDSS telescope on the ground has captured a vast amount of quasar spectra from the early universe. For the first time, a trained deep neural network has discovered weak neutral carbon absorption line probes created by the cold medium of early galaxies within this quasar spectral data. [IMAGE: YI YUECHEN]
An international team led by Professor Ge Jian from the Shanghai Astronomical Observatory of the Chinese Academy of Sciences recently conducted a search for rare weak signals in quasar spectral data released by the Sloan Digital Sky Survey III (SDSS-III) program using deep learning neural networks. The study, published in Monthly Notices of the Royal Astronomical Society, introduces a new method for exploring galaxy formation and evolution, showcasing the broad potential of artificial intelligence in identifying rare weak signals in astronomical big data.
“Neutral carbon absorbers” from cold gas with dust in the universe serve as crucial probes for studying galaxy formation and evolution. However, the signals of neutral carbon absorption lines are weak and extremely rare. Traditionally, astronomers have struggled to detect these absorbers in massive quasar spectral datasets using conventional correlation methods. “It’s like looking for a needle in a haystack, requiring significant effort, time, and inefficiency,” Professor Ge explained. It was not until 2015 that a research team discovered 66 neutral carbon absorbers in the spectra of tens of thousands of quasars released earlier by SDSS, marking the largest sample obtained at that time.
Professor Ge’s team designed and trained deep neural networks with a large number of simulated samples of neutral carbon absorption lines based on actual observations. Applying these well-trained neural networks to the SDSS-III data, the team quickly discovered 107 extremely rare neutral carbon absorbers. This number nearly doubles the largest previously obtained sample, having successfully detected more faint signals than before.
By stacking the spectra of numerous neutral carbon absorbers, the research team significantly enhanced the ability to detect the abundance of various elements and directly measure metal loss in gas caused by dust. The results indicate that these early galaxies, containing neutral carbon absorber probes, had undergone rapid physical and chemical evolution when the universe was only about 3 billion years old (the current age of the universe is 13.8 billion). These galaxies were entering a state of evolution between the Large Magellanic Cloud (LMC) and the Milky Way (MW), producing a substantial amount of metals, some of which bonded to form dust particles, leading to the observed effect of dust reddening.
This discovery independently corroborates recent findings by the James Webb Space Telescope (JWST), which detected diamond-like carbon dust in the earliest stars in the universe, suggesting that some galaxies evolve much faster than previously expected and challenging existing models of galaxy formation and evolution.
“Unlike the JWST, which conducts research through galaxy emission spectra, this study investigated early galaxies by observing the absorption spectra of quasars,” Professor Ge noted. He emphasized that applying well-trained neural networks to find neutral carbon absorbers provides a new tool for future research on the early evolution of the universe and galaxies, complementing the JWST’s research methods.
Professor Ge stated “To uncover more ‘treasures’ in massive astronomical data using artificial intelligence, it is necessary to develop innovative artificial intelligence (AI) algorithms that can quickly, accurately, and comprehensively explore those rare and weak signals that traditional methods struggle to find.”
The team aims to further promote the innovative methods used in this study, extending them to image recognition by extracting multiple related structures to create artificial “multi-structure” images for efficient training and detection of faint image signals. “The methods of artificial intelligence deep learning hold enormous application value and potential in multi-domain image recognition and weak signal detection,” Professor Ge envisaged, “In the future, we hope to uncover more treasures from massive astronomical data.”
For more information, please contact:
Professor Ge Jian
E-mail: jge@shao.ac.cn
Shanghai Astronomical Observatory,
Chinese Academy of Sciences
Source: Shanghai Astronomical Observatory,
Chinese Academy of Sciences