Collaborative Research: SLINGSHOT: Decoding Dark Matter through Gravitational Lensing

点击翻译按钮获取中文摘要

The effects of dark matter can be seen in gravitationally lensed systems where light from a distant galaxy is gravitationally lensed by a foreground galaxy. Analysis of the resulting distorted, magnified, multiple images of the background galaxy provide clues to the nature of dark matter. The principal investigators (PIs) Gleyzer, Ames and Alexander will develop the Strong Lensing Inference and Generative System for Harnessing Observational Telescope Data (SLINGSHOT) framework for simulating realistic gravitational lensing images and applying machine learning algorithms to identify known categories of possible dark matter substructure. The PIs will broaden participation with a series of online lectures on machine learning and astrophysics as part of the National Society of Black Physicists Computational Physics Initiative. The team will also continue to teach data science concepts with Jupyter notebooks through their Creative Open Data Environment (CODER) education and outreach project for K-12 teachers and students.The PIs will establish a baseline for classification of dark matter substructure morphology using both conventional (Bayesian) and machine learning algorithms. They will train the machine learning algorithms in anomaly detection, classification and property estimation using a library of representative dark matter models. Unsupervised anomaly detection techniques will identify promising candidate images in a model-independent manner and localize anomalies for further analysis. The PIs will use the code to analyze gravitational lensing data from the Hubble Space Telescope and the Hyper Suprime-Cam on the Subaru Telescope, in preparation to apply their methods to survey data from the Vera C. Rubin Observatory. The SLINGSHOT toolkit will be made available to the astronomical community.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.