Job Description

Two PhD student positions open. Two examples of potential research directions are briefly outlined below:

1. Development of robust machine learning methods:
Machine learning (ML) methods leverage past training data to perform tasks that either assist or automate human decision-making. Most methods build on the assumption that data distributions from a training environment and future test environments will match each other. However, when these environments diverge, the task performance of state-of-the-art learning methods will degrade and may result in erroneous decisions. New methods that are robust to such divergences are needed in a wide range of applications, including wireless receiver localization, health diagnostics, and medical decision support. We seek to develop such methods through a cross-fertilization of ideas and tools from different branches of machine learning, statistics and optimization.

2. Developing flexible models for nonlinear dynamics:
Dynamical models are essential in explaining the world around us. When it comes to modeling nonlinear dynamics, flexible models have the potential to significantly enhance performance over simple ones. A key topic here is to design deep models with longer memories. Related to this is the task of deriving algorithms that can be used to learn the unknown model parameters from the measured data. If there is interest we will also look into applications within life science and medicine.

Follow @eurasip