My experience at Sheffield has definitely helped with my career

A student in the peak district national park
Katerina Margatina
PhD Student
Computer Science, Natural Language Processing research group
Katerina is doing her PhD in Data Efficient NLP with Active Learning. She recently started the third year of her PhD and this year will be doing two internships, one in Amazon and one in Meta (ex-Facebook).

Can you tell us more about your background?

I was born and raised in Athens, Greece, where I completed my undergraduate studies in Electrical and Computer Engineering at the National Technical University of Athens (NTUA).

Why did you choose the Department of Computer Science at the University of Sheffield for your post-graduate research?

The Computer Science department at the University of Sheffield has had traditionally a very strong NLP group (among the best in EU, not only in the UK), so it attracts exceptional students and academics. Still, the main reason I chose to come to Sheffield is my main supervisor, Nikos Aletras. He is an amazing researcher in NLP, he had an Amazon Alexa scholarship to fund my PhD, and in the interview he seemed like a cool person to work with!

What do you particularly enjoy about studying Computer Science?

I have had the privilege to study a wide spectrum of Engineering fields in my undergraduate studies; from electronics to automatic control, telecommunication to energy systems. I realised that what I liked the most in Computer Science is Natural Language Processing (NLP). It's just fascinating to try to "teach" human language to machines!

Tell us about being a postgraduate student here. What’s a typical week like for you?

A typical week as a PhD student here would involve going to the office at Regent Court 3-4 times per week (1-2 days I work from home). When at work we often eat lunch together with my colleagues, sometimes at a restaurant close to the department. Our recent post-Covid tradition is to go to the pub after work every other Friday (and we have a lot of fun)!

What is your favourite thing about Sheffield?

Sheffield strikes the perfect balance between a big city and a small town; not as hectic as London, but still there are many things to do. My favourite thing would be "the Peaks", i.e. the option to go to the Peak District during the weekend for hiking and visiting the beautiful villages. Being in touch with nature is quite important to me.

Can you tell us a bit more about your project and what you are working on?

I am working on Natural Language Processing (NLP) and Machine Learning (ML). The ultimate goal is to explore the limits of language technology in the sense of how we can build models that help us in our everyday life (consider search engines, translation models, etc). Until now, such models are based in supervised learning, that is training models that aim to 'mimic' human behaviour and generalise to new situations. To do that, we need supervised data, meaning textual data that are annotated by humans. Annotation could be, for instance, the label 'positive' or 'negative' in a tweet (and this NLP task would be Sentiment Analysis). This is where my part is. I aim to research which (unlabeled) data are the most informative or useful for annotation! Under the assumption that 'not all data is equal', the field of Active Learning aims to find iterative algorithms that select the most informative data, using models-in-the-loop.

What are your plans for the future and how do you think your experience at Sheffield will help you in your career?

I recently started the 3rd year of my PhD, so I can start seeing the end! I am excited because I will be doing two internships this year, one in Amazon and one in Meta (ex-Facebook). I think it is amazing that in our field (NLP/ML) we have the opportunity to work for a few months for the best tech companies in the world and gain invaluable experience. My experience at Sheffield has definitely helped with my career, as I've learned how to work hard, publish papers to top tier conferences and, finally, how to socialise and make new acquaintances. I am particularly grateful to my supervisors for helping me shape these skills, as they are crucial for a successful future career in academia or in industry. Finally, it is great that the department supports such initiatives (like internships), by letting the students apply for a placement or leave of absence, and consequently granting an extension (of the respective months) to submit their PhD.

Any tips for someone thinking of doing a PhD?

The most important advice in my opinion is to prioritise the relationship with one's supervisor. I think the reason I feel great about my PhD is due to the constant support from my supervisors, Nikos and Loïc. They have been there for me in highs and lows, pushing me when I needed to be pushed and supporting me when I wanted time off, always prioritising my mental health and wellbeing. It is great to feel close to one's supervisor, constantly balancing between considering them as a 'friend' and a 'boss', but never really one or the other. My tips to new PhD students would be to find a supervisor that they communicate well with and that they feel they can built a relationship based in trust and honesty. It is also useful to have research interests that align very well, but from my experience this is not super important. It's a bonus if you can achieve pluralism in supervision, meaning that when you have two (or more) supervisors you can get advice and opinions from different perspectives -- which is very beneficial for creativity and inspiration in research!

Find a PhD

Search for PhD opportunities at Sheffield and be part of our world-leading research.