What will you be doing?
In the department Intelligent Imaging we use deep learning algorithms to solve computer vision tasks. Deep learning algorithms are very popular in this field because of their superior performance compared to traditional machine learning methods, but they come at the expense of requiring large amounts of training data. For a new prestigious DARPA research program ‘Learning with Less Labels’, we are investigating new ways to train neural networks on less data. The ultimate goal is to train on an order 106 fewer samples compared to common approaches.
A very promising approach to using fewer samples is to use domain adaptation. A model is optimized for the training samples, which, in domain adaptation, differs from the test domain. For such a different domain, the environment may look different and even the objects may look different. Perhaps there are even different object classes. Often a learned model does not perform well for such cases. Typically, a few of samples of the test domain are labelled and model parameters are updated to behave well on the new samples. This is known as domain transfer. But sometimes the labelling of new samples is hard or even impossible. For instance, when the object of interest is rare.
We consider the problem of learning a new object with very few or even no labels. Often, much is known about the new object. For instance, when searching a new kind of car, it is known that it has wheels, windows, doors, and side mirrors. The goal of this internship is to exploit such world knowledge. For instance, by modelling the relations between its parts, learned from images of similar objects. We believe that graph networks are able to exploit semantic relations to embed world knowledge. The ultimate goal is to improve domain adaptation by utilizing world knowledge.
What do we require of you?
- You are in the final stages of your degree in artificial intelligence, computer science, physics, mathematics, electrical engineering or a similar degree and are familiar with deep learning and computer vision.
- You have experience in programming in Python. Knowledge of Keras is preferred.
- The internship will be starting in August 2020.
What can you expect of your work situation?
TNO is an independent research organization whose expertise and research make an important contribution to the competitiveness of companies and organizations, to the economy and to the quality of society as a whole. Innovation with purpose is what TNO stands for. With 3000 people we develop knowledge not for its own sake but for practical application. To create new products that make life more pleasant and valuable and help companies innovate. To find creative answers to the questions posed by society. We work for a variety of customers: governments, the SME sector, large companies, service providers and non-governmental organizations. Working together on new knowledge, better products and clear recommendations for policy and processes. In everything we do, impact is the key. Our product and process innovations and recommendations are only worth something if our customers can use them to boost their competitiveness.
The Intelligent Imaging research group is a young, passionate and creative group of professionals (40 people) specialized in the development of groundbreaking applications in the fields of computer vision and deep learning. One third of the group has started working for us in the past four years.
What can TNO offer you?
You want to work on the precursor of your career; a work placement gives you an opportunity to take a good look at your prospective future employer. TNO goes a step further. It’s not just looking that interests us; you and your knowledge are essential to our innovation. That’s why we attach a great deal of value to your personal and professional development. You will, of course, be properly supervised during your work placement and be given the scope for you to get the best out of yourself. Naturally, we provide suitable work placement compensation.
For this vacancy it is required that the AIVD issues a security clearance after conducting a security screening. Please visit for more information the AIVD website.
Has this vacancy sparked your interest?
Then please feel free to apply on this vacancy! For further questions don’t hesitate to contact us.
Note that applications via email and third party applications are not taken into consideration.
Contact: Wouter Uijens
Phone number: +31 (0)6-158 87391