Identifying and Characterizing Strong Gravitational Lensing Systems Using Probabilistic Machine Learning

Gravitational lensing occurs when a massive object, such as a galaxy cluster, bends the fabric of spacetime, turning it into a magnifying glass that focuses the light from a distant object behind it. This phenomenon allows astronomers to observe faint astronomical objects and some of the earliest faraway galaxies or black holes that would otherwise be invisible and probe the distribution of matter in the “lenses” themselves. During this interdisciplinary research project, the intern will combine technical skills (computer science) to (1) compare machine learning models, (2) optimize them to help detect these “lenses”, and (3) incorporate scientific (astronomy) knowledge to analyze practical usages of the identified images and calculate some of the lenses’ physical properties (e.g., the distribution of dark matter). This project is especially relevant today, with huge amounts of high-quality images of the large-scale structure of the Universe being taken by the recent Euclid mission that will be impossible for us to search through without the use of automated algorithms. These strong lensing systems should also allow us to derive new measurements of the expansion rate of the Universe (the Hubble constant), a quantity whose value is the subject of intense debate among astronomers today.

Faculty Supervisor:

Joshua Speagle

Student:

Partner:

National University of Kyiv-Mohyla Academy

Discipline:

Physics

Sector:

Artificial Intelligence; Other; Aerospace

University:

University of Toronto

Program: