Why designing computers more like brains is a smart decision


9 Nov 2021

Image: Debismita Dutta

The energy-saving advantages of neuromorphic computing were explained in just three minutes at FameLab Ireland.

Debismita Dutta, a chemical engineer from India pursuing her PhD at Tyndall National Institute, had no idea what to expect when she received an email about participating in a science communication competition called FameLab.

“This looks cool,” she thought. “Might as well register.”

Dutta was new to Ireland at the time and saw this as an opportunity to be productive outside the lab, where she is researching magnetic nanomaterials with a focus on data storage devices.

“It is only after participating that I found out the scale at which this event takes place and the long legacy it holds in the science communication space,” she told Siliconrepublic.com.

FameLab challenges participants to deliver a talk on a scientific subject in just three minutes, and Dutta said the learning experience of the contest was “immense”. Applying her knowledge of the world’s advanced computing needs, she focused her talk on the drive to build computers more like brains.

‘For neuromorphic computing research to gain steam, we need to propagate its importance to people’
DEBISMITA DUTTA

What inspired you to become a researcher?

My mother used to buy me a lot of science encyclopaedias and mechanical building sets growing up. Even though I might not have understood much at the time, I did know that I enjoyed knowing more about things as I was a very curious kid.

Given that science was an interest that I acquired over time, it made sense to become a researcher and make that very ‘knowing more’ gig my career.

How was your experience with FameLab?

Science is epic and has played a huge role in our evolution. But unfortunately doing impactful science requires funding, and that can only happen when the general public knows about all the exciting things science can bring them. I think FameLab aids that very well.

I am glad I registered that day because this experience has taught me a lot about how I’m going to present my science to others convincingly. Even hearing other people’s talks made me realise that we can do so much with how we package our research for conferences and lectures.

How would you sum up your FameLab presentation?

My talk is about how computers are taking up too much energy, and with the way the world’s computation needs are increasing, we wouldn’t be able to conjure up the required energy to power our computers in the future.

The alternative lies in mimicking the design of our brain and applying it to computers as our brains are highly energy-efficient and powerful computers. This new age of computing is called neuromorphic computing.

Why did you choose neuromorphic computing for your presentation?

With everything around us becoming smart and programmable, we are entering a new age of technology. An age where everything from medicine to agriculture or culinary science requires the support of algorithms to stay relevant in the new innovation market.

This also means that we will be increasing our energy intake by a huge amount. The current system is highly inefficient and wouldn’t be able to keep up with our needs in the future. Hence, it needs a major upgrade. For neuromorphic computing research to gain steam, we need to propagate its importance to people. I figured this talk is a good way to do so.

What is the biggest challenge you’ve encountered in science communication?

With science communication it is important to remember that the audience is not necessarily well versed in your subject, and it is important to dilute the vocabulary into small digestible pieces that are easy to grasp and understand. I have personally found it quite challenging since I am prone to using verbose sentences that might not be very interesting to someone hearing about a concept for the first time.

There is a teaching aspect to this, and understanding that has helped me overcome it. I remember that some of my teachers presented very difficult topics in a very broken-down format with the help of metaphors and examples. This not only helped me understand the concept better, but also piqued my interest in learning more about the subject. That is what I have tried to imitate when I communicate my own science, and I wish my audience to have a similar takeaway as I did.

What common misconceptions about science would you like to correct?

Since school we have always heard that there are three states of matter. Wrong. There are actually four.

Solid, liquid, gas and … drumroll … plasma.

If you want to see what that is, microwave a grape right now. (You can read the rest later. Seeing plasma formation in action is more important.)

Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.