When classifying high-dimensional sequence data, traditional methods (e.g., HMMs, CRFs) may require large amounts of training data to avoid overfitting. In such cases dimensional...
We face the problem of novelty detection from stream data, that is, the identification of new or unknown situations in an ordered sequence of objects which arrive on-line, at cons...
This paper presents a robust unsupervised learning approach for detection of anomalies in patterns of human behavior using multi-modal smart environment sensor data. We model the ...
We have developed a method that can discriminate anomalous image sequences for more efficiently utilizing security videos. To match the wide popularity of security cameras, the me...
We consider a semi-supervised regression setting where we have temporal sequences of partially labeled data, under the assumption that the labels should vary slowly along a sequen...