Three-Toed Sloth

Upcoming Gigs: Newell-Simon Hall

Next week and the week after, I will attempt to convey everything in this class and my research which is relevant to machine learning in two one-hour talks on stochastic processes, over lunch.

"Stochastic Processes and Their Prediction", Monday, 5 November 2007, 12 noon in 1507 Newell-Simon Hall, Carnegie Mellon University

Abstract: Stochastic processes are collections of interdependent random variables; this talk will be an overview of some of the main concepts, and ways in which they might interest people in machine learning. After a brief mathematical introduction, I focus on stochastic processes whose variables are indexed by time, which are closely related to dynamical systems. The key problem here is understanding the dependence of the variables across time, and the different sorts of long-run behavior to which it can give rise. I will talk about various kinds of dependence structure, especially Markov dependence; how to give Markovian representations of non-Markovian processes; and how to use these Markovian representations for prediction. Finally, I'll close with some recent work on discovering predictive Markovian representations from time series.

"TBD", Monday, 12 November 2007, 12 noon in 1507 Newell-Simon Hall

Abstract: TBD.

If I am good, the hour next week will actually be enough to cover the material in the abstract, and the week after I'll talk about stochatic processes in space. If not, I'll finish up. (In other words: lectures will continue until morale improves.)

Enigmas of Chance; Self-Centered