Ecole Polytechnique (France)
MILA (montreal institute for learning algorithms)/ Element AI
Dphill student in Oxford.
Favourite thing to do in research: Exploring
Dphill (i.e. young researcher) in mathematics in Oxford
Hello! I am Jacob, originally from France but now doing my research in Oxford university. I am 24 years old, have 8 little brothers and sisters, and love mathematics and artificial intelligence: those two words are mysterious/ scary as my friends sometimes tell me. Also the job and lifestyle of a researcher are somewhat unusual. I believe one can be convinced that being a researcher in Science is cool for the two following reasons:
- A researcher works/explores cutting edge topics, which he is the only one to understand completely. These subjects, sometimes, have the potential to change the world.
- A researcher is free to choose the subject of his work and study. He can choose also to teach other people, choose to not work on Mondays, choose to take holidays when he wants but work during week ends. Briefly, he can choose his lifestyle without too much constraints from the outside.
Let me introduce myself now. Below is a (random) picture of me:
I was 15/16 years old at that time, with no idea of what I wanted my career to look like. But what was clear is that I already took pleasure in solving mathematical problems. In fact, I also love to teach to others (obviously, having 8 little brothers and sisters I developed somehow a teaching methodology). For instance, a few years back, I taught mathematics and physics to 120 chinese students in Beijing, for 6 months. Let me show you one picture of me at that time (in my office, on the left, with my students in class, on the right). Exercise: find me in the middle of the class on the right picture.
Let me tell you about my favorite hobby now: I love travelling. As soon as I spot cheap train tickets/ a week day or any opportunity to travel, I try to seize it. It is always a wonderful experience to be confronted to different people and culture right? I travel with friends more often than not, and come back with unforgettable stories. Examples of places I loved: Brazil/ Japan/ Italy/ China/ Russia/ Peru/ Iceland/ Israel. As a proof of my trip in Brazil, let me show you a picture of an unbelievable landscape of Lencois de Maranheses (never know how to spell this one):
Looking forward to meet some of you!
Researcher in mathematics and Artificial intelligence
My work focus on diverse topics.
Artificial intelligence is growing fast
The scary word/world artificial intelligence does not mean anything, because it encompasses a very wide range of very different topics. The general idea of an artificial intelligence, also called A.I. is that of a robot/ or a piece of code in a computer, that is able to do a task that human do in every day life. For instance, there is an artificial intelligence, called Convolutional Neural Network (abbreviated CNN), which is able to recognise people/ features in images and videos. An other one, called Recurrent neural networks (RNN), can understand text and speech. Sometimes A.I can even do what human can’t. For instance, we may see soon autonomous cars, those that do not need drivers. You may also know another famous example, the one of an A.I. able to beat human in the Go game. Recently in fact, a new A.I. has been able to defeat humans at Poker.
I love Generative Adversarial Networks (GAN), a specific type of artificial intelligence. Let me tell you what a GAN can do. Give a GAN a bunch of pictures of yourself, say 100 pictures, some in class, others at home, or during holidays, with friends or family, etc. Then the GAN will analyse these pictures and generate infinitely many new pictures that look very realistic but do not exist. Here are two funny ways in which GANs have been used so far:
- GANs can create paints that look very modern and realistic. In fact, humans are unable to guess whether a given picture was created by a GAN or another human. Look at this paintWell, whether it is beautiful or not is hard to decide (taste and colors right?). However, this GAN-created paint has been sold 432 500 dollars in auctions.
- GANs can also create trendy music. Skygge is a band which consist in no human. All voices and sounds are GAN-created. The following link is an example of song by Skygge https://www.youtube.com/watch?v=iuWYQe3aGlg
A.I. has many other impacts. For instance I worked also at Owkin for a few months, a wonderful startup that uses A.I. to solve medical questions. They develop models that are able to detect cancers for instance.
Mathematics is an art also
I also do mathematics. To me, problems in mathematics are fascinating. Let me give you a few problems (Don’t try to solve any of them, they are hard):
- Is there a set that contains all sets?
- Is there something that is either true or wrong, that we can prove to be unprovable?
Some thousands of other problems do exist and are still not proved. Sometimes people even think a problem is solve and we happen to realise centuries later that there was a mistake in the proof. The examples above are somewhat abstract, but maths is also about solving concrete and realistic problems:
- You are in front of two doors, one leads to a treasure of an eternal life and the other to a disgusting soup you will have to eat (made of toothpaste, beetles and a bit of olive oil). You don’t know which door leads to what. But there are two guards in front of the doors. One is telling the truth always, the other always lie. Unfortunately you don’t know which guard lies and which one tells the truth. You have the right to ask a unique question that can be answered by “yes” or “no”, to one of the guard (the other won’t answer). What is the question you ask in order to know which door leads to the eternal life?
More generally, maths is about writing clearly problems arising in everyday life. It appeared a while ago. When people realises that trunks of trees could roll, they wanted to have the concept of circle so as to design wheels. They had to make a mathematical definition of the circle, which is nothing but the abstract following definition: a circle is a set of points which are all at the same distance from one points. Beautiful right? Anyway, mathematics are almost as old as humankind, so it is not surprising that we today have thousands of problems to solve.
The great advantage of mathematics, in my opinion, is that they are purely conceptual and do not need any reality other than the mind to exist. Said in a more down-to-earth way, to do biology/physic or other sciences, you need to setup experiments, and confront your observations to reality. On the opposite, mathematics do not need reality, as mathematical objects are purely abstract and invented by ourselves. In that respect mathematics looks like any other art: it can be created and developped with few constraints from the real world.
The topic I study is called Persistent Homology, and it is great fun! It consists in explaining mathematically what are the differences between distinct shapes. For instance, a difference between an “8”, a “0” and a “1” is that there is 2 holes in the “8”, 1 hole in the “0” and none in the “1”. Persistent homology is about calculating the number of holes hidden in shapes for more complicated shapes.
My Typical Day
Going to the Maths institute in Oxford and enjoying the work atmoshphere
I wake up when I want -8AM/9AM- and go to the maths department of Oxford: a wonderful place with a few hundreds of researchers like me. I spend a lot of time working in my office, where there is a computer I can use to read articles on the internet, and a board to write my thoughts.
They are seminars happening all the time: researchers gather together for a certain amount of time to discuss their works. Often groups of people find an interesting problem and try to solve it together. I particularly like collaborating this way.
I also teach a few classes to undergraduate students, and can also decide to pop up in classes as a volunteering student if I want to learn about a specific topic.
What I'd do with the money
Open to suggestions
How would you describe yourself in 3 words?
Passionate/ Responsive/ Young
What's the best thing you've done as a researcher?
I do not know
What or who inspired you to become a researcher?
My former teachers
What was your favourite subject at school?
What did you want to be after you left school?
Were you ever in trouble at school?
Yes and No
If you weren't a researcher, what would you be?
Who is your favourite singer or band?
What's your favourite food?
Burrata (some specific mozzarella cheese)
If you had 3 wishes for yourself what would they be? - be honest!
Find wonderful results in my research