Machine Learning (ML)
Many of the most visible advances in the field of Data Science all require some form of machine learning techniques. Machine learning is a central topic of interest to DKE, with applications spanning multiple research themes.
Machine learning is a sub-field of Artificial Intelligence that aims to develop self-improving computer systems and discover the fundamental laws of learning. The field develops ways for computers to learn how to perform tasks, through trial and error or by having the machine look at examples and working out how to do it by itself. This contrasts conventional ways of programming, which mostly present pre-defined sets of instructions for a computer to follow.
Machine learning requires knowledge of calculus, linear algebra, statistics, optimization and so on. It is – among others – involved in research on knowledge representation, search and games, robotics, and multi-agent systems. DKE is either involved or has expertise in all of these fields. As such, machine learning is a topic of interest across the department as a whole.
Research focus and application
Due to its central role in Data Science, machine learning plays a role in a significant number of DKE research projects. Notable examples include applications such as breast cancer classification, predicting drug-induced gene expression changes, human visual cortex activity interpretation, life quality improvement for the elderly, study selection help, e-coaching, and so forth.
In addition to these application areas, we also develop new techniques for recommender systems, reliable predictions, human-brain-based techniques for deep learning, kernel based deep learning methods, etc.
Machine Learning in practiceDr. Stelios Asteriadis on improving quality of life for patients with dementia through deep learning |

Scope
- Prediction and classification
- Applications and understanding of Deep Learning and recommender systems
- Transfer Learning and Domain Adaptation
- Text Mining
- Reliable Prediction
- Recommender Systems
- Applications and Representation aspects of (Multi-Agent) Reinforcement Learning
- Evolutionary Computation
The ML website is under construction: ML was previously embedded in the Robots, Agents and Interaction (RAI) group.
Go to RAI website
Researchers
Highlighted publications
- Ismailoglu, F., Cavill, R., Smirnov, E., Zhou, S., Collins, P., & Peeters, R. (2020). Heterogeneous Domain Adaptation for IHC Classification of Breast Cancer Subtypes. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 17(1), 347-353. https://doi.org/10.1109/TCBB.2018.2877755More information about this publication
- Mehrkanoon, S. (2019). Deep neural-kernel blocks. Neural Networks, 116, 46-55. https://doi.org/10.1016/j.neunet.2019.03.011More information about this publication
- Zhou, S., Smirnov, E., Schoenmakers, G., & Peeters, R. (2017). Conformity-Based Source Subset Selection for Instance Transfer. Neurocomputing, 258, 41-51. https://doi.org/10.1016/j.neucom.2016.11.071More information about this publication
- Mocanu, D. C., Bou Ammar, H., Lowet, D., Driessens, K., Liotta, A., Weiss, G., & Tuyls, K. (2015). Factored four way conditional restricted Boltzmann machines for activity recognition. Pattern Recognition Letters, 66, 100-108. https://doi.org/10.1016/j.patrec.2015.01.013More information about this publication
Dotti, D., Popa, M., Asteriadis, S. (2019) A Hierarchical Autoencoder Learning Model for Path Prediction and Abnormality Detection. Pattern Recognition Letters. https://doi.org/10.1016/j.patrec.2019.06.030
Roefs, A., Boh, B., Spanakis, G., Nederkoorn, C., Lemmens, L., Jansen, A. (2019) Food craving in daily life: comparison of overweight and normal‐weight participants with ecological momentary assessment. Journal of Human Nutrition and Dietetics. https://www.doi.org/10.1111/jhn.12693