---

Research Interests: Foundations of Information Theory 

Shannon gave a compelling, principled measure for the mutual information that two random share. But how do we measure information shared amoung more than two variable?

E. Schneidman, S. Still, M. J. Berry II, and W. Bialek. Network Information and Connected Correlations. Physical Review Letters, 91(23):238701, 2003.

Student Projects (499, 699) and thesis projects:

1. Information provided by a measurement
            If we learn a parametric model, the information provided by an experiment can be measured by the information gained about the (unkown) parameters. But what is the approporiate measure for non-parametric models. In this project we study the existing approaches in the litereature and come up with new ideas.

2. Non-stationary processes
            Real world learning problems are often non-stationary. What are reasonable measures of uncertainty and information in non-stationary processes?

---