Outreach: The Rise of Artificial Intelligence in Astronomy

BMSIS Young Scientist Sumeet Kulkami gave a talk at the British Council’s “Cafe Scientifique.” Read his reflections below.

I started with a brief history of the development of Artificial intelligence, and highlighted the difference in how a regular computer algorithm works as opposed to a machine
learning algorithm, using the popular Towers of Hanoi toy. I also talked about how the development of computing power and AI was best reflected in how they conquered two games: chess and GO respectively. The first part of my talk was based on an article I wrote last year on the same theme.

For the latter part, I showed examples of Machine Learning ideas applied in research in Astronomy. Recent successful applications involve classification of different galaxy structures, finding new exoplanets in Kepler data, finding gravitational lenses and ML-based tools to remove noise and clean up astronomical images. There was an enthusiastic question and discussion session afterwards, with the audience interested in talking about some ethical aspects of the research too. One question stood out for being fun: “Are we living in a simulation as Elon Musk says we might be?” To which I replied “Sure, but the simulation will never allow us to find the right answer to that question.”