Projects
Deep neural networks as a model of speech perception KU Leuven
Tensors and Neural Networks for Computational Creativity KU Leuven
Creativity in language is ubiquitous. It is abundantly present in work with an explicit creative intention - such as literary novels or poems - but weighty doses of creativity also pervade everyday language use. We believe that a computational model of creativity that focuses on language will shed light on the enigmatic processes and interactions that come into play when we humans express ourselves in creative ways. Moreover, natural language ...
Miniaturization Effects and Node Placement for Neural Decoding in EEG Sensor Networks KU Leuven
Electroencephalography or EEG can be used in a wide range of applications like diagnosing disorders like epilepsy, Brain Computer Interfaces or in neural prostheses. These mentioned applications would benefit from a continuous 24/7 EEG monitoring, but this is impractical with traditional bulky EEG headsets. On the other hand, the recent emergence of novel miniature and wearable EEG devices could make it possible, yet these devices come with ...
Using Model-Based Reinforcement Learning combined with Monte-Carlo Tree Search to optimize Neural Networks for Embedded Devices. University of Antwerp
Modelling and understanding aesthetic preferences for visual patterns, photographs and paintings: Comparing human perceivers with convolutional neural networks KU Leuven
In spite of the wide-spread belief that “beauty is in the eye of the beholder”, recent research in empirical aesthetics has focused on the role of statistical image properties as quasi-universal, biologically rooted factors underlying the preference for some patterns, photographs, and paintings. In the slipstream of the booming area of machine learning (deep neural networks, DNNs, and convolutional neural networks, CNNs), a new field has ...
Computational Modeling of Social Cognition and associated Deficits by means of Artificial Neural Networks KU Leuven
Anomaly Detection in Spatiotemporal Neural Networks KU Leuven
In this doctoral study, we study and refine techniques to find anomalies in spatiotemporal image data from various industrial applications. After all, we notice that better detection can be achieved if the frame-by-frame processing of an image sequence is abandoned and the video is processed as a coherent entity by the neural network. This is true for applications of object detection, but certainly for action recognition is the inclusion of ...
Adaptive hardware for conditional neural networks with emerging technologies KU Leuven
The work revolves around the investigation of architectures for integrated systems that enable training and inference of neural networks. Particular attention is reserved to the realization of hardware that adapts its performances (in terms of power, delay and resource usage) depending on the context of utilization. Emerging technologies and materials for memory/interconnections are studied as well with the aim of improving the performances ...
Relating Auditory EEG to Continuous Speech Using Artificial Neural Networks KU Leuven
Understanding how the human brain processes speech is crucial for developing tools that can help people with hearing loss or speech related disorders. The electroencephalogram (EEG) is a popular way to measure brain activity in response to continuous speech due to its high temporal resolution and affordability. A common approach involves recording EEG signals from people while they listen to continuous speech, and developing models to relate ...