A few weeks ago, the day was sizzling hot in Chennai and Mercury was soaring high. The TiE (The Indus Entrepreneurs) sponsored event at IITMR had a catchy theme AI.NEXT with one of the topics titled “Julia in Machine Learning”. Who is Julia? What does she do in Machine Learning? The battle between Mercury and Venus, obviously swayed towards the latter and I decided to attend the late afternoon event.

Sukumar Rajagopal (Founder, Tiny Magic) who is closely associated with Chennai Mathematical Institute (CMI), set the context for the three topics viz. “Julia in Machine Learning” by Prof. Sourish Das (Associate Professor, CMI), “ChatGPT for Enterprise” by Suchitra Karunakaran (CTO, Algolabs) and “Future of AI” by Prof. Madhavan Mukund (Director, CMI)

Sourish stunned the audience when he opened with the statement that during the course of his 25 minute orientation, he would make us fall for Julia 😊 I was thinking that with so many people, how would Julia make the choice – will she be using AI/ML for the swayamvar? That Sourish did a commendable job would be evident in the following paragraphs!

What we came to know next was even more shocking, that Julia was born in 2012 and is not even a teen !? Those of you who were expecting a Mills and Boon type story – Sorry, I have to disappoint you! Julia is one of the recent programming languages and has as its siblings the illustrious Python, R etc. The unending stream of programming languages that keep emerging is reminiscent of the unlimited starters that keep getting served in popular barbeque restaurants!

Coming to the crux of the matter, Julia boasts a series of advantages viz. is fast, interactive and easy to learn. Apparently, people who have been experimenting indicate that Julia can offer speeds several times more than Python and can be just as fast as optimized C++ or Fortran (see article in reference section).

Julia avoids “the two language problem” – a very typical situation in scientific computing where a researcher devises an algorithm or a solution to tackle a desired problem or analysis at hand. Then, the solution is prototyped in an easy to code language.

People struggle to overcome obesity and embark on plethora of diet regimes and exercises to fight it out. The world of programming is no different, with bulky programs becoming complex and difficult to maintain. Here is the solution from Julia – it requires several lines less of code, making it much more compact.

The easier syntax and the interactive nature of the language makes the learning and development cycle faster.

And now, whether everything is hunky dory with Julia? Not exactly. Remember that Julia is yet to be a teen. The ecosystem offered by Python or R has evolved greatly over couple of decades and have a significant edge over Julia, at this point of time. But the gap is expected to narrow down in the next 4-5 years.

So, do you love Julia? 😊

References

https://www.matecdev.com/posts/julia-worth-learning.html

https://julialang.org/

https://scientificcoder.com/how-to-solve-the-two-language-problem

https://blog.quantinsti.com/julia-programming/