Academic

Postdoc and PhD positions in MACHINE LEARNING with applications to PROACTIVE INFORMATION RETRIEVAL AND BIOINFORMATICS

Helsinki University of Technology

We are looking for two postdocs and one PhD student to join a research group working on machine learning and probabilistic modeling, with applications in bioinformatics, neuroinformatics and information retrieval. The Machine Learning and Bioinformatics Group (http://www.cis.hut.fi/projects/mi) belongs to the Adaptive Informatics Research Centre and the Helsinki Institute for Information Technology HIIT, Helsinki University of Technology, Finland.

Postdoc #1 and PhD student: Proactive information retrieval

Proactive or cognitive user interfaces require modeling user behavior and state, in order to provide optimal selection of natural choices for the user in varying contexts. They provide an exciting opportunity for developing advanced probabilistic machine learning methods for inferring the user's interest from noisy signals like measurements of attention. You will take part in international projects developing e.g. gaze-based information retrieval tools and personalized information navigators.

Keywords: Proactive interfaces, information retrieval, user modeling, probabilistic modeling, graphical models

Postdoc #2: Bioinformatics

The aim is to develop models that learn to combine multiple data sources in multiple experimental settings, for systems-level modeling of a new experiment. From machine-learning perspective this is a challenging combination of multi-view and multi-task learning, where the number of samples compared to the dimensionality is very low and fundamental research is needed. From systems biology perspective this is an opportunity to solve some of the central problems in cellular regulation. The project is carried out in collaboration with experts on the data and the biological applications (cancer and plant signaling). You will need particularly skills on machine learning and probabilistic modeling, as well as good collaboration skills.

Keywords: graphical models, MCMC, relational models, interactomics, multi-view learning, multi-task learning

More details about applications at http://www.cis.hut.fi/projects/mi/positions08.txt

Samuel Kaski, samuel.kaski@tkk.fi Professor of Computer Science