Summer School Spotlight: Kartik Kumar Tokala
Tell us a little bit about yourself
Currently I am pursuing an MSc in computer science at the University of Nottingham. Before starting my masters, I worked for two years at an IT firm in India (where I’m from) and before that I did a Bachelors in Technology in Computer Science, also in India.
In terms of a research focus, I’m especially interested in Artificial Intelligence but I need to narrow it down to a particular domain as it is such a vast subject. This is something that I’m experimenting with at the moment and is one of the reasons I applied to join this summer school.
Away from studying, I’m interested in sketching anime characters and cartoons. I also like football although I haven’t followed it for a while. I used to play football for my University when I was studying for my undergraduate degree and I also played at high school.
Why did you apply to join the DC Labs summer school?
One of main reasons I applied to join was because of the opportunity to focus on solving a real life problem through research.
I haven’t had much experience in implementing something from scratch for the end user. The DC Labs summer school attracted me because it provided this opportunity.
Another reason that I joined was to gain experience in a field that is new to me – neural networks. The project I’m working on is focused on neural networks, a part of AI that comes under the umbrella of deep learning. This is a popular concept right now, with a lot of research taking place and I wanted to get involved and gain some experience.
Tell us about the project you’re working on
The project I’m working on is focusing on finding traffic patterns in developing countries using a neural network.
Most developed countries have the infrastructure (e.g. cameras, sensors) to find out the velocity of vehicles to predict the flow of traffic and accident hotspots.
In developing areas, infrastructure is non-existent or limited which makes predicting traffic very difficult.
In these areas, drone footage could be used to identify vehicle speeds and density, transport modes, and the interactions of different vehicle modes.
This will enable a cheaper and more flexible approach to traffic analysis and management without requiring any new traffic infrastructure, enabling identification of hotspots for predicting accidents.
What would you like to achieve this summer?
Over the summer I’d like to gain more experience training a neural network. I’ve previously undertaken small scale projects but I’ve never implemented anything like this.
Whilst I’m in touch with my supervisors on a regular basis, this project requires me to work independently. I’m enjoying trying to figure things out myself and search for solutions. I think that this will help me to gain research skills.
In the future I would like to pursue a PhD, but at present I’m exploring domains. I think this summer will be really helpful in helping me explore a new field and will feed into my thinking about a research question for a PhD.
And what are your first impressions of DC Labs?
This is the first time I’ve worked as an intern and my early impressions are very good ones.
After my Bachelors in Technology I went onto work for an IT firm and so I do have prior work experience, but being able to work on something that I’m really interested in, and that is in line with my study goals, is a great opportunity.
The programme includes a good mix of people and an amalgamation of ideas and disciplines. It’s well-structured and there are opportunities to generate lots of ideas.
It’s also been interesting to hear about the other projects, such as designing a game interface using GANs.
I think that I’ll be able to use my experience over this summer for many good reasons in the future.