Erik Rehn

Learning representations that disentangle hidden explanatory factors in data has proven beneficial for effective pattern classification. Slow feature analysis (SFA) is a nonlinear dimensionality reduction technique that provides a useful representation for classification if the training data is sequential and transitions between classes are rare. T...

Direct gaze is a potent non-verbal signal that establishes a communicative connection between two individuals, setting the course for further interactions. Although consciously perceived faces with direct gaze have been shown to capture attention, it is unknown whether an attentional preference for these socially meaningful stimuli exists even in t...

Hierarchical temporal memory (HTM) is a biologically inspired framework that can be used to learn invariant representations of patterns in a wide range of applications. Classical HTM learning is mainly unsupervised, and once training is completed, the network structure is frozen, thus making further training (i.e., incremental learning) quite criti...

Hierarchical Temporal Memory is a biologically-inspired framework that can be used to learn invariant representations of patterns. Classical HTM learning is mainly unsupervised and once training is completed the network structure is frozen, thus making further training quite critical. In this paper we develop a novel technique for HTM (incremental)...

Capacitive sensing is used in many different fields of application. It has been implemented in such devices as mobile phones and remote controls. However, up until now the physical sensing area has remained limited despite the widespread use of larger input devices such as keyboards. We present DGTS, which seamlessly integrates keyboard typing and...