The Ups and Downs of a Graduate Student

I have been dreaming about doing research since I started as a bachelor student in computer science. The idea of doing original research was really mysterious and exciting at the same time. I didn’t do any research at that period of my life (something that I would consistently regret), so the idea was still vague to me. I did some quick research but it was only in mathematics and published a 200 page on advanced integration. That strengthened my background in mathematics but I would struggle to apply that in my field. Given that, I was still looking forward to doing more research and this time in computer science.

As excited, I started my masters in computer science and I would still wonder around the different topics to see the topics that interest me the most. Still, I would move from one topic to another lost between the things that interest me and those that seem popular in the literature. At that time, deep learning was capturing a lot of hype but I lacked all the necessary requirements (no background whatsoever). I spent a whole semester reading about different papers in the field. I was under the pressure of starting my thesis topic in computer vision but was stressed about the idea of skipping some necessary concepts. I would say, that period was the most self-doubting period in my life as a student. A lot of expectation with little to advance. You know when you expect doing something novel in the field and end up doing way below the state of the art. Maybe the problem was going in a direction where a lot of researchers had spent a lot of time and experience in and I am just starting and expecting to break the numbers. I came to know that this is usual as a starting graduate student but the idea of getting sota results was still intriguing to me. Maybe I was supposed to think about something fundamental rather than challenging numbers.

After three years of struggling I finished my masters degree with a lot of lessons that I learnt. I started my PhD with much more confidence and a little more narrowed research. I completely switched from doing computer vision to NLP. This improved my confidence a lot. Moreover, collaborating with others has benefited me a lot instead of just battling thoughts in my mind. Discussing ideas with colleagues was really beneficial to me to improve my research experience. Still, I would question myself and whether I am making the correct decisions. The idea of creating something novel still confuses me and whether I can do it or not. Working on different things has benefited me a lot and gave me the motive to succeed in many different endeavours. I am currently in my second year and hoping to learn more and do original research related to Arabic NLP.