My research interests are in the area of machine learning and on applying learning techniques
to practical tasks.
Specifically, I am interested in new methods for learning probabilistic and causal models for event detection and, in general, for temporal data, learning from users/experts (I am a member of DARPA's Bootstrapped Learning program), semi-supervised learning, learning and decision making on a budget (when decisions cost) and with non-uniform loss functions, the automation of feature construction & selection, active learning, hierarchical learning, and incorportaing domain knowledge into machine learning algorithms.
I am also interested and working on testing and validation methods for
decision systems, learned models, learning algorithms, and large systems that are based on learned models.
In general, I am interested in listening and talking about inductive learning, methods for
scaling up and improving the performance of learning techniques, computational and statistical learning theory, unsupervised and reinforcement learning, game theory.