"Fields of Learning Theory, Machine Learning Grow Together at University"

The University of Chicago recently played host to workshops on learning theory and machine learning, two disciplines that are increasingly intermixing, according to Toyota Technological Institute (TTI)-Chicago professor John Langford. The learning theory workshops involved the sharing of ideas between mathematicians, statisticians, and computer scientists on the mathematical properties of evolutionary processes, and one workshop emphasized the evolution of language. Computer science professor Partha Niyogi says language evolution illustrates the transmission of knowledge through learning instead of inheritance. "You actually have to learn [your parents'] language not just from data provided by them, but also data provided by many other language users in the community in which you are immersed," he explains. The learning theory workshops were organized under a $2.2 million National Science Foundation grant awarded to Niyogi and TTI-Chicago professor Stephen Smale. The two-week-long Machine Learning Summer School involved the participation of around 100 students and represented a collaborative initiative between TTI-Chicago and the university's computer science department. The field of machine learning focuses on teaching a computer to learn from experience to carry out tasks that people can perform, but at substantially less cost. Niyogi says learning theory and machine learning have many mathematical foundations in common, and both the Machine Learning Summer School and the learning theory workshops represent the culmination of the three-month Program in Learning Theory and Related Areas organized by the university's computer science department and TTI-Chicago.

-University of Chicago Chronicle (06/09/05) Vol. 24, No. 18; Koppes, Steve

http://chronicle.uchicago.edu/050609/machinelearning.shtml