Physics-Guided Machine Learning

Description

This project aims to harness the potential of Physics Informed Neural Networks (PINNs) in the study of dark matter, utilizing strong gravitational lensing images. Strong gravitational lensing, a phenomenon predicted by Einstein’s theory of general relativity, occurs when a massive galaxy or cluster of galaxies bends the light from a more distant galaxy into a ring or arc. This effect not only magnifies the background galaxy but also distorts its image, providing a unique opportunity to study the mass distribution of dark matter in the lensing galaxy or cluster.

The project will focus on developing a PINN framework tailored to infer the properties and distribution of dark matter in lensing galaxies. The PINNs will be trained on simulated and observed strong gravitational lensing images, incorporating physical laws governing lensing phenomena directly into the learning process. This approach aims to improve the accuracy and efficiency of dark matter studies by seamlessly integrating physical constraints with deep learning methodologies.

Duration

Total project length: 175/350 hours.

Difficulty level

Advanced

Task ideas

Build various physics informed neural network architectures that are endowed with known physics.
Apply these models to study dark matter in strong lensing images in various contexts: classification, regression, anomaly detection, and more.

Expected results

Demonstrate the superior performance of these architectures with real and simulated data sets.

Requirements

Python, PyTorch and relevant past experience in Machine Learning.
Familiarity with astrophysics and gravitational lensing is preferred but not required.

Test

Please use this link to access the test for this project.

Mentors

Michael Toomey (Massachusetts Institute of Technology)
Sergei Gleyzer (University of Alabama)

Please DO NOT contact mentors directly by email. Instead, please email ml4-sci@cern.ch with Project Title and include your CV and test results. The relevant mentors will then get in touch with you.

Corresponding Project

DEEPLENSE