Machine Learning in Astronomy

Program WS2     List of Participants WS2


The availability of large-scale data in the last two decades has pushed astronomy into a data-intensive field of science rather than a data-limited one. Upcoming large telescopes such as Vera Rubin Observatory, Dark Energy Survey, Square Kilometer Array, etc. (to name a few) will increase the availability of astronomical data by many folds. To handle such a vast amount of data, traditional methods are already falling short, and astronomers need to utilize powerful and highly efficient machine-learning methods which have the potential to fully exploit the exponentially increasing data. Over the last few years, machine learning methods have continued to grow in popularity in all areas of astronomy, promising significant scientific advances. Therefore we feel that the Indian Astronomy community, especially the young astronomers, needs to be exposed to the prospect of machine learning for astronomy.

We wish to propose a workshop on machine learning in astronomy in the upcoming 41st meeting of the Astronomical Society of India (ASI) at IIT-Indore. In this workshop, we aim to introduce machine learning methods to the participants through introductory lectures, hands-on sessions, and short talks by experts who used machine learning for astronomy research in their recent works. The goal of this workshop is (1) to familiarize participants with the basic machine learning techniques, (2) teach them how to write simple algorithms that can benefit their scientific research and, (3) give them a glimpse of current research in Indian Astronomy performed using machine learning.

The tentative plan of the workshop is as follows. It will begin with a couple of introductory lectures by experts in machine learning, followed by a hands-on session. In the hands-on session, python jupyter-notebooks with detailed documentation will be shared with the participants. These jupyter-notebooks will be run on Google’s colab environment that provides GPU processing capabilities. Google colab will minimize the time required for software installation on individual local machines of the participants. In the afternoon session, we aim to familiarize the participants with the ongoing machine-learning efforts in the Indian astronomy community. For that purpose, experts (around 6 to 8 young astronomers) already working in the field will be invited to give short presentations about their recent work. This workshop will help grow awareness of machine learning techniques among Indian Astronomers, motivate young astronomers to try new methods to handle large data and provide an opportunity to collaborate and discuss with experts. We will also make the resources used in the hands-on session publicly available to all the participants of the ASI meeting.

Format of the meeting

  • We expect the participants to be 50 (40 participants + 10 resource persons)
  • For the afternoon session, we will invite a few (6-8; counted as resource persons) young Indian Astronomers who are actively working on machine learning in Astronomy
  • The time-table below provides our tentative plan for the one day workshop
  • 9:00 AM - 09:45 AM Introductory Lecture 1
  • 09:45 AM - 10:30 AM Introductory Lecture 2
  • 10:30 AM - 10:45 AM Tea-break
  • 10:45 AM - 11:15 AM Hands-On : Exercise 1
  • 11:15 AM - 11:45 AM Hands-On : Exercise 2
  • 11:45 AM - 12:15 PM Hands-On : Exercise 3
  • 12:15 PM - 01:00 PM Hands-On : Q&A + Challenge
  • 01:00 PM - 02:00 PM Lunch-break
  • 02:00 PM - 02:45 PM Challenge continued
  • 02:45 PM - 03:45 PM Three case studies of 15+5 mins
  • 03:45 PM - 04:00 PM Tea-break
  • 04:00 PM - 05:00 PM Three case studies of 15+5 mins


Vivek M (IIA), Vikram Khaire (IIST), Ajit Kembhavi (IUCAA)