PhD Positions
Here you find a real life guide through PhD position applications containing useful information of how to find the positions, advice from successful candidates, the real questions asked during the PhD interviews and so on.
This document is always updating. Feel free to check in back in some time. I have published it sooner, so if anyone needs to have a look or know some questions could have it already.
PhD Hunting Guide
An unofficial guide for Cognitive Science master students of CIMeC seeking PhD positions after their master made by actual graduated students.This neat and useful document is the result of kind contributions of: (alphabetically)
Alireza M. Kamelabad (LMI-2018)
Ekaterina Torubarova (CNS-2018)
Eleonora Gualdoni (LMI-2018)
Greta Gandolfi (LMI-2018)
Find Your Fit
One of the key factors of a successful PhD application is to find your fit. It is important that you know your own interests, and also you can find what others are interested in for the position they publish. There are certain ways to find the current available PhD positions to apply for. Here we discuss all these matters.
Find Positions
TBA
Application Materials
In this part we will go through the main PhD application materials and documents that most of the institutions and universities will ask you for. For now, these documents and the tricks are limited for the PhD positions within European Union institutions and countries since the regulations are different from the rest of the world. However contributors can add their experiences.
Curriculum Vitae
TBA
Motivation Letter (Statement of Purpose or Motivation)
TBA
Research Proposal
TBA
Transcripts of Records
TBA
References
TBA
Real PhD Interview Examples
Here you find the real PhD interviews that contributors have done since 2018. It can give you a good understanding and impression of how an interview for a PhD might be and think which kind of answers might be good for you and to prepare in advance. (All the informations and provided materials in this page have received proper consent from the data subject institutions and people.)
Institution
Pompeu Fabra University (UPF)
Barcelona, Spain
Application Result
Successful
Interview Date
April 2020
Research Group
COLT
The general description of the interview ambience
I had an interview with all the team members individually. The project supervisor was super nice, interested in what I want to do more than in what she’d like me to do and tried to understand me instead of putting me in trouble. With the PhDs I had just informal interviews, that the project supervisor wanted me to do to help me in getting a clearer idea on what the team does. Similarly with the postdocs, except for one who was formal and strict (he/she was kind of responsible for the project, together with the project supervisor, so it makes sense).
Advice
I gave general answers. My ideas were not detailed. I just showed which aspects of the project could be interesting for me, giving some details on some related papers I read. I did not have a clear structure for a future project and they were perfectly fine with it.
In my case, the PhD was highly related to the topics I was working on in the thesis. That helped me because they could see a particular fit for the role.
Interview Questions
What are your interests? What really makes your eyes shine?
Which questions would you like to pursue in your phd?
Precise description of the thesis and some clarification questions about it
A lot of "what do you mean by" and some questions about the details of the thesis project. Example: "how are you going to evaluate this phenomenon?"
Some clarification questions about the cover letter
A postdoc was interested in my Bachelor thesis so he/she asked to tell him/her more in detail
Which are the strengths that you'll bring to the group? What are your abilities that you think would be useful and how you learned them?
Given this dataset (they sent me the paper in advance), how would you test the phenomenon X? If your first idea didn’t work, what would you do, then? Here’s what I would do. Do you see problems in this procedure? (All this was by the “strict” postdoc).
Would you be ready to relocate?
Do you like more qualitative/statistical analysis or modeling? What would you like to do in your phd? (Feel free to decide, it’s your phd. This is just for info)
Which is your experience with deep learning? (spoiler: around zero)
Have you applied elsewhere?
Institution
Radboud University - Nijmegen, Netherlands
Position Title
PhD fellowship in Cognitive Computational Neuroscience
Application Result
Unsuccessful
Interview Date
February 2020
Research Group
Private on Request
The general description of the interview ambience
It was rather stressful I would say. There were four people in the interview, two of which had more than 15,000 citations each! The other committee member who actually was the head of the institution was not present there as planned due to personal reasons.
Advice
It was my first application and I messed up the interview, but later I got very good feedback from the responsible professor. I asked her for advice of what I can improve about my application and here is her reply.
Interview Questions
To be added
Pre Interview Tasks
Before the interview I had to write a critical review between two papers that they sent me a week prior to the interview and that was an important part of the selections process. There you should show your knowledge of the field.
Institution
University of Amsterdam (UvA) - Amsterdam, Netherlands
Position Title
4 year PhD position: Computational Linguistics
Application Result
Unsuccessful
Interview Date
April 2020
Research Group
ILLC
The general description of the interview ambience
I was invited to a zoom call 20 days after I sent my application. I joined a single call. The atmosphere was quite formal, we just went through the formal process (which looked exactly like a university oral exam), no chat, no small talk. Prof. AAA and Dr. AAA were nice and calm, while the responsible professor was more serious and firm.
Interview Questions
I was asked to read, summarise and design my contribution to their project proposal. They provided a well designed and interesting project plan 2 days before the interview. So, for the first 10 minutes I was required to give a short presentation of what I thought about.
Clarification questions of the form: ‘You mentioned x,y,z. Can you provide a more detailed explanation of these factors?’ x 3
Do you think that this pipeline (imagined relying on w2v representations) would work if we use BERT? What should we change? In what sense? Be as clear as possible.
Do you think that sentiment analysis can be applied to our project? How?
I’m impressed by your grades. How do you explain them? How do you obtain high grades even if you don’t have a bachelor in maths/engineer/computer science? How much do your grades deviate from the mean of your colleagues? (the latter one was NOT so explicit, but it was what they wanted to know!)
You have an interdisciplinary background. How can it help to deal with our projects?
What skills do you think you will need to improve? What are your strengths?
What do you want to focus on after this PhD (subject-wise)
Why did you apply for a position in our group? Did you know it?
Do you have any questions for us?
Institution
Radboud University - Nijmegen, Netherlands
Position Title
4 years PhD position in "Developing technology for long-term engaging health-promoting chatbots"
Application Result
Unsuccessful
Interview Date
April 2020
Research Group
BSI
The general description of the interview ambience
They have been so nice and friendly. For the first 10 minutes they just introduced themselves and we chatted irrelevantly and informally I introduced myself. Then they asked me to give my presentation. After that they started asking their questions. They mostly asked the questions about my motivation letter and research proposal I had sent them plus about my presentation. There were other questions about my personality and interests.
Advice
I got rejected but I was the second person. Here are the emails which I think help you see the reasons for the decision.
Then I replied them asking for some more feedback about the results:
And they replied to that:
Pre Interview Tasks
I was asked to write a research proposal based on their project proposal for max 800 words.
And also to prepare a presentation of the last research work I have done.
Interview Questions
Present the last research project you have been working on. (Thesis or any related research) [10 minutes]
The relation between this project you have done and our PhD project. In the proposal that you sent you also mentioned that you would like to use active learning in the chatbot project. Please elaborate on this. What are the things you have learned from this project that you can also use for this project?
What attracts you the most in this position?
You are making the project bigger than we thought (It is based on my answer to the previous question and the proposal I had.) You said that the project is realistic but at the same time you are probably also aware of the limitations that chatbots have and you mentioned it yourself in the proposal. How would you for bridging this gap; The limitations that chatbots have at the moment and the ambitions that we all have including you for the next few years?
I really like your presentation and I like the work that your colleagues do that is of course very theoretical of course but this project is more applicational, I was wondering whether you are aware of the differences in the nature of these research? (He meant are you ok with the non-theoretical nature of our project?)
Thank you for the very nice presentation, I am the one who has expertise so a word of a complement. One question about proposal and also our proposal, there is the trust issues with the chatbots, I found it a bit difficult how you are going to address that in terms of your background, I would like to hear your ideas about how you are going to do that, so how the machines through text generation and text comprehension becomes a trust-worthy partner?
Part of the project is human-chatbot relation and another part is human-human so the chatbot listens to the humans' conversation, how do you think of all the text generation will interplay in the conversation of two humans? How would you approach trust in that case? How do you try to incorporate the fact that in one of the projects that we have two people are talking and the chatbot is the third person.
Do you have any existing models in mind for that? (My answer to the previous question that we can use the computational models of dialogue modeling.)
You said you are very motivated to help people stop smoking and of course there are millions of people who could benefit from this application, now what if that’s not working, so how do you deal with the failure? When you're now trying to work on a project for four years?
I liked my answer here, so I share it: Regardless of the results if we can make an application which helps people stop smoking significantly what do we learn the most is that what did what try and what did not work out! So we are aware of the things which do not give us good results, so then we can go for the things which we did not try yet. That is not nice for myself to see my project does not work, any researcher does not want negative results but it is how the science goes on and I just learn from that to know what does not work.
Part of your PhD program is to teach, very small proportion. Do you have any idea which teaching you could do?
In your proposal you said you would like to work with PyTorch, how much data would you need for this kind of approach?
The Questions Asked by Candidate
Can I do teaching? How much of the project requires me to teach?
How is the nature of collaboration and team working in your institution?
Does the institute facilitate the collaboration between me and the other two PhD students working on the project?
Do you facilitate for me to learn the Dutch language?
How much freedom do I have in doing the research itself? Am I able to have my own ideas and integrate them in the project?
How is the supervision system in this project?
Institution
University of Amsterdam, the Netherlands
Position Title
4 years PhD on “Cognitive Neuroscience and Computer Vision”
Application Result
Unsuccessful
Interview Date
July 2020
Research Group
---
The general description of the interview ambience
The ambience was chill and nice because the PI was alone and she stated in the beginning that she wants to just talk and get to know each other. It was her first ever call for PhD students so it seemed that she isn’t sure what to ask and she’s more interested to evaluate my personality and research goals rather than specific experience/skills/technical details. I presented my master thesis in 15 minutes, she thanked me for a nice presentation and asked a couple of questions about which results I expect and how I’m going to analyze the data. Then she asked questions about me, and let me ask questions from her. In total the interview took 1.5 hours.
Advice
Prepare abundant answers to at least two questions: 1) why do you want to have a PhD; 2) why do you want to do it in this exact place, why did you choose this project out of all. If you don’t have direct experience with something, at least show how eager you are to learn it, or that you indirectly touched it through something else. Try to not leave any questions unanswered with just ‘No’.
Go through the publications / collaborations of the PI or the whole selection committee, read 1-2 papers which seem the most interesting, mention that you have read them and liked them. I got some answers to detailed questions by skimming through a couple of papers by the PI before the interview. Try to find some connections between you and the PI (maybe you know someone who they worked with or some place where they have been).
Either prepare a good lie or be honest, but don’t say unprepared lies to look better!
Interview Questions
Describe how your ideal supervisor should be and what you expect from communication with supervisor
Why do you want to have a PhD
Do you have an image about what PhD work consists of and what responsibilities you will have
Where else did you apply and why did you consider the Netherlands
Do you like writing / presenting / working with participants
Did you work with cluster / Unix / github
Did you have experience with computer vision software and neural networks
Do you know what is Convolutional Neural Network
Do you have any science heroes in your field
How can you transfer your knowledge about language into vision
Institution
Furhat Robotics, Stockholm, Sweden / Manchester University, Manchester, UK
Position Title
3 years PhD on “ESR14: Child-personalized adaptive language teaching through interactive social robot”
Application Result
Successful
Interview Date
July 2020
Research Group
---