Dr. Anima Anandkumar is a principal scientist at Amazon Web Services and a Bren professor in the CMS Department at CalTech. She is the brain behind Amazon SageMaker, an end-to-end Machine Learning integration application. Her research interests are in the areas of large-scale machine learning, non-convex optimization and high-dimensional statistics. In particular, she has been spearheading research on the development and analysis of tensor algorithms. She is the recipient of several awards such as the Alfred. P. Sloan Fellowship, Microsoft Faculty Fellowship, Google Research Award, ARO and AFOSR Young Investigator Awards, NSF Career Award, Early Career Excellence in Research Award at UCI, Best Thesis Award from the ACM Sigmetrics society, IBM Fran Allen PhD fellowship, and several best paper awards. She has been featured in a number of forums such as YourStory, Quora Sessions on Machine Learning, O’Reilly Media etc. She received her B.Tech in Electrical Engineering from IIT Madras in 2004 and her PhD from Cornell University in 2009. She was a postdoctoral researcher at MIT from 2009 to 2010, a visiting researcher at Microsoft Research New England in 2012 and 2014, and an assistant professor at U.C. Irvine between 2010 and 2016.
Dr. Anima Anandkumar is a highly distinguished alumnus of our very own institute, who now rules the world of AI and machine learning. She mentioned that her affliction towards math and engineering began at a very early age. Over her fireside chat, as a part of the Shaastra 2018 Spotlight Lecture Series, she told us how Feynman’s lectures inspired her early love for physics and how she has always felt that math is a natural language to converse. It is thus that she has been spearheading the development and analysis of tensor algorithms with such apparent ease, which she believes will be the big leap in space efficiency for machine learning routines.
During her talk, she spoke of her days at IIT Madras with great nostalgia. When introduced to the audience as an alumnus of IIT Madras, she proudly added that she isn’t just an alumnus of IIT Madras but an alumnus of the Shaastra Core Committee as well. She said that the transition from introducing guest speakers during her organising days to being one of the speakers at Shaastra has indeed been special.
Her Shaastra Spotlight lecture was a treat for all machine learning enthusiasts. In the two hours of her talk, the audience was injected with the most potent dose of technical knowledge.
From your undergraduate days at IIT Madras to being a Principal AI Scientist at Amazon, you’ve surely had an inspiring journey. When did your passion for research begin?
Since my earliest childhood, I have always loved math. The order and beauty of math just spoke to me. Reading stories about Albert Einstein, Richard Feynman; how they discovered things that changed the world had an impact on me, I’d also watch a lot of documentaries about science and the discoveries that can be made. Hence, from my very early days, I was fascinated with doing and learning new things; and that has enabled me to be a researcher.
Despite coming from a small town in India, you’ve conquered all stepping stones and challenges to rule the world of AI, which is now undeniably one of the hottest research areas. What is your advice for women trying to make their way in the tech world?
There are so many societal challenges that we still need to overcome today. Apart from the existing setback in terms of opportunities, girls are still faced with a social perception which convinces them that mathematics and science are not their forte. I was very lucky because my mother is an engineer and my grandmother, even though not formally educated was very passionate about puzzles and mathematics. I had very strong women role models growing up and I think that there has to be some awareness starting from the family that women are equally capable and should be given opportunities.
Right now, it’s beginning to happen. There are high school and college-level camps for women. There are seminars where only women scientists present their findings. Such gatherings are extremely crucial as they build a community of women in technology and enables us women scientists to realise our combined potential.
What is your take on the proposed 14% reservation of seats for women at IITs? What do you think are better suggestions to improve the skewed gender ratio and encourage girls to pursue science?
Even though I missed having more women in IIT, the women who got in there were remarkable since they overcame other barriers and still performed well; it gave a lot of confidence. Though I do wish there were more women and I’m always looking how to improve the diversity, it should be towards helping women overcome barriers without compromising on performance/quality. I think that the proposed move would only hamper the confidence of women.
Anima Anandkumar had started a petition on change.org against the proposed reservation for women at IITs.
You’ve been sharing your time between Amazon and CalTech. What are your viewpoints on creating a truly collaborative world between academia and industry?
It has indeed been a very exciting journey over the last year, having roles at both Amazon and CalTech. I found support from both the sides and we were able to forge a partnership between Amazon and CalTech, where there is a collaboration award of cloud credits to CalTech by AWS. I believe such a collaboration helps bridge the barrier and provide more computing infrastructure to universities. I look forward to a lot of intellectual exchange between the two communities.
Personally for me, going to AWS has been a huge learning experience, seeing developers build production quality code. It has been a very different and much-needed, exciting academic exposure.
You have followed unconventional teaching methods to make classrooms more interactive and application oriented. What are your ideas for making classes more engaging and less burdensome?
Having huge classes is not the way. Using AI and more online tools to give students individual attention and adapt to their individual pace of learning is very effective. There’s this concept of the reverse classroom where there are no lectures at all and classes are completely interactive. This extreme view might discourage students to speak up. Hence, I do lectures and then have interactive question and answer sessions, where students can select their answer to the question posed and view the responses of the entire class immediately after. So, if there are trick questions or common mistakes, they can see that it’s not just them but a significant chunk of the class who got it wrong. Being able to adapt the latest technology into education is crucial.
What is your advice to budding academicians trying to come up with something original? What would you say to students trying to get started on research?
Even when you’re learning in your courses, be very critical, and don’t just accept anything. When presented with a very efficient method, question yourself, ‘Is it really? When will this fail?’ Try to be adversarial. Try to play the devil’s advocate. Try to understand or figure out the limitations of the currently existing methods and try to think of ways to conquer them. Always look at the challenges and find out if people have already thought of a way to solve the problem. I think that these will help you go to the next level.
