The Security for Machine Learning group (SfML) investigates attacks on Machine Learning algorithms and countermeasures to protect against them. We are looking for a candidate who can help us with the implementation of attacks and countermeasures. A strong background in the development of software is therefore desirable.
Good mathematical insight is required; knowledge or experience with Machine Learning is a plus. We mainly investigate Convolutional Neural Networks.
Your work will consist of implementing countermeasures, setting up and running experiments and processing / analyzing the results. The follow-up steps will be determined based on the results. Alignment therefore happens daily.
The software we use to perform and verify attacks is written in Python (using Keras/Tensorflow) and runs on both Windows and Linux (Ubuntu). The heavy experiments are performed on AWS instances with GPUs. All experiments are logged in a SQL Database. For version management, we use Git with a basic branching set-up where pull requests to master are reviewed. For all these techniques, practical experience is required.
The leading semiconductor supplier for the secure identification, automotive and digital networking industries. Our top client enables secure connections and infrastructure for a smarter world, advancing solutions that make lives easier, better, and safer. As the world leader in secure connectivity solutions for embedded applications, our client is driving innovation in the secure connected vehicle, end-to-end security, and privacy and smart connected solutions markets.
Uren per week: 40