| |||||||
Application Keynote Lecture
Prof. Adam Amara (University of Surrey)
Applying AI to Frontier Science Experiments
AbstractObservational cosmology, the study of the Universe, is going through a golden age. In particular, we are in the midst of an influx of data from ongoing experiments, such as the Dark Energy Survey (DES). In the coming years, the volume and quality of data will rapidly increase as mega-surveys such as Euclid, Rubin, and Roman come online. Processing these data will require new algorithms and methods to maximise our scientific reach and control for errors. In this new era of massive astronomical data, new AI and machine-learning techniques offer exciting opportunities to explore the universe and uncover new discoveries about fundamental physics.In this talk, I will discuss the challenges of applying machine learning methods to high-precision science experiments. Given the complexities of the late-time Universe, these approaches must capture the important properties of galaxy populations and key features imprinted on the experiments' data. By combining all these elements with advanced statistical methods and new machine learning algorithms, we can build a process for extracting maximal information from the new data to account robustly for error effects. I will conclude with lessons learned for applying AI to other areas of science where precision, reproducibility and robustness are vital.
Adam Amara is a research scientist who studies the universe's evolution and the structures that lead to galaxies. This area of science has many unanswered questions that challenge our fundamental understanding of physics. They include trying to understand the mysterious effects of dark matter and dark energy, which are not sci-fi terms but measured phenomena that dominate the universe's evolution.
| |||||||
