Current opportunities:
MSc: MSc Computational Neuroscience, Cognition and AI (started Fall 2019)
There are regular openings for PhD students in Computational Neuroscience. Feel free to email me. Follow this link to make a formal application: PhD opportunities
Frequently asked questions
I am often asked by physicists, computer scientists, biologists, or mathematicians about studying computational neuroscience. Here are personally biased answers to some of those questions.
Q: I want to switch to computational neuroscience. What do I need? How should I prepare for an MSc?
First, you might feel more secure knowing that many people before you have made this switch. Clear scientific thinking is the most important skill. Next, I think, are quantitative skills. Typical tools I use for my work are: statistical physics, non-linear dynamics and simulations. Programming is an important secondary skill. However, it is important to point out that excellent programming by itself does not make good science. Programming languages that are used a lot in this field are: Python and Matlab.
While biology is highly complex, most of it can be learned by self study as you go along. For our MSc, no prior biology or psychology knowledge is required.
Q: I already have a PhD. Should I first obtain a PhD in computational neuroscience before I can work in this field, or can I go for a postdoc position straight away?
Not necessarily. A large part of the skills you learned during your PhD (writing papers, scientific thinking, independent work) will be also useful when you switch fields. However, there is a danger that you will not get exposed to enough neuroscience. You should consider this when looking for postdoc positions. You need to be extra pro-active to study neuroscience and to integrate into your new field.
Q: Do I have to do experiments?
No. Some researchers working in neural computation never see a lab. Others are captivated by studying the nervous system directly and turn into full-time experimentalists.
Q: What attracts you in Neural Computation?
The nature of my personal fascination with neuroscience is not very different than my fascination with physics. Physics tries to figure out the rules governing the physical world and tries to describe the world in mathematical terms. This seems a crazy idea, but the weird thing is, of course, that it works. The realization that all that we see is made out of atoms, that everything obeys the same laws, is awesome. In neuroscience one tries to understand how the brain works, and therefore what the basis is for our thinking. And, crazy as this idea might seem, in some cases we start to understand it. We know roughly how the sensory systems and the motor systems work, especially for the lower animals. The realization that all our thinking is done by those things called neurons is in my opinion equally astounding as relativity theory. What makes neural computation particularly fun is that many questions remain still open and can be researched. The field also moves very fast, so that a newcomer in the field has an advantage. Q: What else should I do to decide on an area of interest? Talk to people in the field. Conferences are a very good source of information, not only for experts. Read a few books. A starting student can easily asks question to which nobody knows the answer. What is even better, the student can try to learn the necessary techniques and explore the issue for themselves after a couple of months.
Q: How much maths do I need, and how can I read up on it?
Computational modelling builds on maths. The more maths you know, the better.
There is a number of excellent Mathematics for Engineers textbooks. Important topics are Linear Algebra, Series Expansion, Differential Equations. Less important are Laplace Transform, Complex Variables, Special functions, Trigonometry, Vector Calculus.
Reading list computational neuroscience:
- "The Computational Brain" by Patricia S. Churchland, Terrence J. Sejnowski. MIT Press; ISBN: 0262531208; Reprint edition (February 3, 1994). Gentle Introduction - "Theoretical Neuroscience : Computational and Mathematical Modeling of Neural Systems" by Peter Dayan, L. F. Abbott. MIT Press; ISBN: 0262041995;
- "Biophysics of Computation: Information Processing in Single Neurons" by Christof Koch, Oxford University Press; ISBN: 0195104919. Focusses on single cell modelling
Reading list AI and neural networks
- "Introduction to the theory of neural computation" Hertz, J., Krogh, A., and Palmer, R. G. (1991). Perseus, Reading, MA. [Technical, but very thorough book.]
- "Understanding Deep Learning" Simon Prince. Mit Press.
Online lectures:
Current Research
Mark van Rossum started his computational neuroscience research after obtaining a PhD in theoretical physics. He has since worked across a wide number of topics in neuroscience, including, retinal function, synaptic plasticity and homeostasis, memory formation, noise in neurons and sensory coding. In addition, he has developed a number of data analysis methods for neuroscience, including the 'Van Rossum' spike distance measure.
For his research he mainly uses computer simulations and theoretical analysis, and of course extensive conversations with fellow scientists. He is always interested in setting up new collaborations.
Current research includes the study of schema and rule extraction from memory events, and the role of metabolic constraints on neural computation.
The most up-to-date list of publications can be found on Google Scholar.