Aras Selvi

About you

Why did you decide to study the Doctoral programme at Imperial College Business School and what makes this programme unique?

Firstly, I was drawn to London. As a Business School student, it is very important for me to be engaged with businesses during my PhD years. Needless to say, London is a vital hub in the global business world, and Imperial College Business School has a great business network. Secondly, what makes Imperial special for me is its science and engineering focus, approached in an interdisciplinary manner alongside the main business studies. With a background in engineering and applied mathematics, my research centres on optimisation and machine learning. Besides being a PhD student at the Business School, I am also part of the Computational Optimisation Group in the Department of Computing, and affiliated with the Data Science Institute and The Sargent Centre for Process Systems Engineering.

“What makes Imperial special for me, is its science and engineering focus, approached in an interdisciplinary manner alongside the main business studies.”

Finally, and crucially, my PhD supervisor, Professor Wolfram Wiesemann, greatly influenced my decision. Before applying to Imperial, I was reading his papers. He's a Professor of Analytics and Operations at the Business School and also heads the department. His work in 'Distributionally Robust Optimisation' impressed me, and I wanted to join his research group.

What is your previous academic and work experience and how did this prepare you for the programme?

My journey here was a series of interconnected experiences, each one naturally leading to the next: I majored in Industrial Engineering with a minor in Computer Science, which equipped me with the tools to formulate and solve problems in the industry, ranging from supply chain management to forecasting. I undertook two internships at Coca-Cola Icecek, where I worked on developing decision-making tools for optimisation of sugar procurement strategy under uncertainty. My internships were my first experience with decision-making under uncertainty, and I was so intrigued by it that I wanted to learn more about the theory behind it. This led me to do a two-year master programme in Operations Research at Tilburg University, focusing on robust optimisation under the supervision of Professor Dick den Hertog. During my master's studies, I read the work of my future PhD advisor, Professor Wiesemann, and had the opportunity to meet him when he visited our department. Our discussions about exploring new applications of robust optimisation in machine learning inspired me to draft an application letter to Imperial College Business School, ultimately leading to the start of my PhD journey here!

The programme

How did the MRes prepare you for your doctoral research?

At Imperial, our initial year is dedicated to a Master of Research (MRes) programme, which is specifically designed to support our PhD research endeavours. A significant advantage of this programme is its flexibility, allowing students the freedom to select from a wide range of modules. Additionally, it allows for the enrolment in modules offered by other departments. In collaboration with our supervisors and department managers, we carefully choose the best combination of modules that align with our research objectives. My personal selection included three modules from the Electrical and Electronic Engineering department, two from the Operations group within the Business School, one from the Business Analytics programme, and additional modules from the Mathematical Finance group.

What area of research did you do during your PhD?

My research interests are the theory of data-driven decision making under uncertainty and its applications in machine learning, privacy, and fairness. In my recent works, I have been working on designing optimal privacy mechanisms, developing efficient algorithms for robust machine learning, as well as approximating hard decision-making problems via robust optimisation. Currently, I am researching how robust optimisation can aid in learning from synthetic data. With rising privacy concerns, big companies are increasingly sharing synthetic versions of real datasets. Consequently, researchers are developing algorithms to discern the relationship between this synthetic data and the underlying truth, and how to apply decisions derived from synthetic data to real-world scenarios. My current work focuses on exploring how robust optimisation could enhance the current state-of-the-art in this field.

Which seminars, events or guest lectures at the school have been useful in developing your skills and knowledge?

Our department hosts regular seminars featuring researchers from our field. These events include presentations of their work and meetings with faculty and PhD students, providing an excellent opportunity to connect with esteemed faculty.

We also benefit from access to seminar series offered by the Computing, Statistics, Machine Learning, and Systems Engineering groups. Furthermore, PhD students have the privilege to organise their own seminars and can participate in mock seminars to prepare for future job talks, especially beneficial for those aiming to remain in academia.

How would you describe the Business School faculty and your fellow Doctoral students?

Our faculty in the Analytics & Operations Group is diverse, with professors engaged in both theoretical and applied research. The group includes both experienced senior professors, who are editors at respected Business journals, and a dynamic group of younger faculty members, with several new hires last year. They have strong international connections, having earned their PhDs from renowned universities worldwide, which provides us with extensive networking opportunities.

Our student cohort is equally diverse. There are four of us on the programme. I focus on decision-making under uncertainty; another works on dynamic pricing; a third specialises in revenue management and statistical learning; and the fourth is dedicated to healthcare operations. All of us aim to pursue academic research careers, fostering a supportive environment where we share advice and encouragement.

Opportunities

Have you benefited from being part of the wider Imperial College London community?

The supportive environment at Imperial meant that I received a great deal of help and encouragement. An alumnus introduced me to JP Morgan AI Research's synthetic data group, which led to an internship after discovering we were a great match. Furthermore, I participated in the Enrichment Scheme at the Alan Turing Institute, drawing on advice from Imperial alumni who had previously completed the programme to enhance my application. I was also invited to both give and attend seminars across various Imperial research groups, enjoying discussions and pizza events, which broadened my network and initiated collaborations on a wide array of topics. 

“As a PhD student at Imperial, you become part of a vast community that transcends individual disciplines, fostering a sense of belonging to a large, supportive family.”

What are your future career goals and how have they been realised since being at Imperial?

I aspire to pursue a tenure-track position in academia after completing my PhD. To support this goal, it is important for me to connect with faculty from different universities and gain insights into their institutional visions and research directions. In addition to attending international conferences and engaging in overseas university visits for research collaborations, I also co-founded 'LondOPS', a network for London-based PhD students specialising in Operations Management or Operations Research. Imperial provided a budget for LondOPS meetings, which we hosted on our campus. These meetings have been very helpful in starting collaborations with PhD students from other institutions and do mock job talks.

Advice to prospective PhD students

Every research group has its own unique culture, and I strongly advise checking if you fit well with the culture of the group you're applying to. Additionally, consider the city where the university is located because it often influences the research culture. It's easy to get caught up in the specifics of project descriptions, but remember that you'll also be relocating to a new city and living there for a while. So, take the time to explore the city and see how the research group you want to join adapts to its urban surroundings.