In the first virtual astronomical agent or broker which will process data from the NASA-funded network of four “Atlas” telescopes has become the Chilean project ALeRCE (Machine learning for rapid event classification-Machine Learning for Rapid Event Classification), created in 2017 in collaboration between the Millennium Institute of Astrophysics (MAS) and the Center for Mathematical Modeling (CMM), to later join the D foundationata Observatory (DO)the University of Design (UdeC) and the great university. In addition, more than 20 domestic and foreign institutions have contributed to it since its inception.
Contribution of broker
With the integration, recently announced in Ireland, a new phase begins for the project. Atlas considers two telescopes located in Hawaii for the northern hemisphere with one in Chile and another in South Africa for the southern hemisphere, and they have the function of early detection and warning of asteroids that may pose a risk to people’s lives. To achieve this, the four major telescopes that is Atlas observe the night sky four times a night, detecting millions of variable events in the process, which means a lot of data to analyze.
This is precisely what allows ALeRCE, which uses artificial intelligence algorithms and techniques such as machine learning or machine learning with big data or data intelligence to quickly identify and classify objects and events in the Universe, and thus alert the international community very early on to those of greatest scientific interest. The broker Chile works with a hybrid infrastructure, processing data mainly in the cloud (AWS-DO), in the National High Performance Computing Laboratory (Nlhpc) and using its own resources installed in the National University Network (Reuna), a point critical for the exchange of national and international academic traffic.
“Convert this information delivered by Atlas into a data stream that is cross-checked in real time with observations from the Zwicky Transient Facility (ZTF) in California, USA, becoming the first broker that combines large data streams into one system global multi-telescope», underlines about the work of the interdisciplinary team of the ALeRCE within the framework of the integration into Atlas the doctor Francois Forster, principal investigator of the Chilean project and member of MAS and CMM. “This combination of data will improve scientific discoveries and increase the resilience of ALeRCE as an automatic classification system for variable objects in the Universe.“, he argues.
“For years, we have been developing the ZTF-focused AI tools. This new step presents us with new challenges in extending these methods to data from different telescopes. It’s like teaching the computer to learn from different sources, just like us humans.“explains the doctor. Guillaume Cabrerawho is the leader of the zone of machine learning de ALeRCE, director of the Data Science Unit at UdeC and researcher at MAS.
Cabrera says that, since 2019, ALeRCE pmanages approximately 300,000 variable events per night coming from the ZTF observatory, with a dedicated engineering team developing tools for a community of over 6,000 users in 125 countries. Besides, andn one year, it processed more than 200 million alerts in real time, including 40 million images, more than 6,000 supernovae, 60,000 supermassive black holes and 800,000 variable stars.
With such capacity and indicators, integration as the first broker of Atlas adds to several milestones that have highlighted the national project and, above all, the human team after its exploitation.
And it is thatn 2021 was retained as the only broker Chilean among the 7 in the world who will classify the more than 10 million events per night that will be reported by the Vera C. Rubin Observatory, which is under construction in northern Chile to start operating in 2024 and revolutionize astronomical sciences and the understanding of the Universe by combining the tools of the discipline with those of big data. In addition, ALeRCE was the first broker by making their rankings public using artificial intelligence; it is the one that reports the greatest number and the fastest of supernova explosions; and is the first to automatically enrich data with historical information on billions of objects.
In this sense, Dr. Andrew Jordan, director of the MAS and Chief of Science Officer of the Data Observatory (DO), adds that “the greatest dThe challenges for the ALeRCE in the future are to adapt to handle the huge volume of data coming from the Rubin observatory, to integrate data from more telescopes and to develop new artificial intelligence tools. that allow the international community to derive the best scientific benefit from these observations. In particular, a major challenge is develop tools to detect these objects of unknown nature, that could revolutionize our understanding of the dynamic Universe over the next decade.”
More information on http://alerce.science/.
#UdeC #part #broker #process #data #Atlas #telescope #network