Postdoctoral Position in Machine Learning for Radio Astronomy

Postdoctoral Position in Machine Learning for Radio Astronomy 
Department of Physics, Université de Montréal & MILA
Application Deadline: 1 may 2025
Start Date: Summer or fall 2025

The Astrophysics Research Group at the Université de Montréal, in collaboration with MILA (Quebec AI Institute), invites applications for a postdoctoral position focused on developing a cutting-edge machine learning algorithm to tackle the deconvolution challenges in radio astronomy. The successful candidate will be co-supervised by Prof. Julie Hlavacek-Larrondo (primary supervisor) and Prof. Sarath Chandar (MILA and Polytechnique), and will have access to MILA’s world-class research facilities and expertise.

For more information on Prof. Julie Hlavacek-Larrondo research group, please visit : https://www.astro.umontreal.ca/x-tra/

This position is part of a grant awarded through the prestigious IVADO Exploratory Grants program, which supports innovative and high-risk interdisciplinary research.

This is a 1-year position with the possibility of extension depending on funding and performance.

Project Overview

The Square Kilometre Array Observatory (SKAO), currently under construction, is expected to produce over 1 Terabyte of data per second — a transformational step for astrophysics. However, the most computationally intensive challenge lies in the deconvolution of radio images to remove blurring effects caused by the telescope’s point spread function. Traditional methods like the CLEAN algorithm have become computationally prohibitive and struggle with complex or extended radio structures.

This project aims to revolutionize radio data processing by developing a new algorithm based on Transformer neural network architectures. 

Key Responsibilities

• Lead the development and implementation of the algorithm.
• Compare the algorithm's performance with traditional CLEAN methods.
• Collaborate closely with experts at MILA and within the international SKAO network.
• Publish research results in high-impact journals and present findings at international conferences.

Qualifications

• A PhD in astrophysics, physics, computer science, or a related field (completed or near completion).
• Background in machine learning and data analysis.
• Experience with Python and machine learning frameworks (e.g., TensorFlow, PyTorch).
• Familiarity with radio astronomy data and interferometry is a strong asset.
• Ability to work independently and collaboratively in an interdisciplinary environment.