Students, experts from around globe flocking to Chicago for programs on learning, evolution, language and computers
The University of Chicago News Office
May 15, 2005 Press Contact: Steve Koppes
(773) 702-8366

Students, experts from around globe flocking to Chicago for programs on learning, evolution, language and computers


Machine Learning Summer School

Program on Learning Theory


Approximately 100 students from across the country and around the world will convene at the University of Chicago’s International House from Monday, May 16 to Friday, May 27 for a crash course in machine learning, a branch of computer science that lends itself to applications as wide ranging as detecting credit card fraud, filtering unwanted e-mail and processing genetic data.

The students will be attending the Machine Learning Summer School, which is being held in North America for the first time since it was launched in 2002 by researchers in Europe and Australia. The summer school is a collaborative effort between the Toyota Technological Institute at Chicago and the University of Chicago’s Computer Science Department. A branch of Japan’s Toyota Technological Institute, TTI-Chicago specializes in graduate instruction and basic research in computer science.

“It’s unique, at least inside the U.S., as far as getting experts from a lot of different areas to come to the same place and talk to the same students,” said summer school organizer John Langford, a research assistant professor at TTI-Chicago. “That’s attractive. Since you have all these people spread across so many different universities, it’s really hard to get a good, overall understanding of their field.”

The Machine Learning Summer School coincides with two weeklong workshops at TTI-Chicago that will bring in approximately 50 experts to exchange ideas with local specialists in the related field of learning theory, especially as it pertains to the evolution of language. Both machine learning and learning theory share many common mathematical foundations, said Partha Niyogi, a Professor in Computer Science and Statistics at the University of Chicago.

Both events—the Machine Learning Summer School and the workshops in learning theory—culminate a three-month Program in Learning Theory and Related Areas organized by researchers at TTI-Chicago and the University of Chicago Computer Science Department.

Machine learning is a growing segment of artificial intelligence that involves teaching a computer to learn from experience to perform tasks that a human could not do, or that a human could do, but it would be less expensive to let a machine do the work. A typical example from the business world would be teaching a computer to detect credit card fraud.

“It’s very difficult to detect fraud as it happens because you’re seeing millions of transactions per day. You can’t really have a human sort through them very well,” Langford said. But a computer can be rapidly taught to detect fraud using far more examples than a human could ever become familiar with in a lifetime.

“The alternative to machine learning is trying to engineer things,” Langford said. But sometimes that is a difficult task. In trying to filter out unwanted e-mail, for example, it is difficult for a computer scientist to write an effective software program for filtering spam.

“The spammers adapt to whatever systems you put up. You need to have a system that can adapt on its own in order to keep up with the spammers,” Langford said.

Machine learning also can be applied to solve problems in biology. For example, the sequencing of the human genome has identified which genes contain instructions for making proteins. But the shape of a protein plays a major role in defining its function in a cell. Machine learning can help biologists use computers to solve the difficult problem of determining the shape of proteins.

The Program in Learning Theory and Related Areas that began in March grew out of one of the first major collaborations between the University of Chicago Computer Science Department and TTI-Chicago. Even before TTI-Chicago formally opened its offices on the University of Chicago campus in September 2003, Niyogi and Stephen Smale, a professor at TTI-C, received a $2.2 million grant from the National Science Foundation to study learning theory. They organized the special program and workshops on learning theory under the auspices of that grant.

Like machine learning, learning theory stands at the crossroads of multiple disciplines. The learning theory workshop from Monday, May 16 to Friday, May 20 will bring together statisticians, mathematicians and computer scientists to discuss mathematical aspects of learning theory. A central focus will be the relation between statistical learning and a subfield of applied mathematics and computer science called function approximation.

The learning theory workshop from Monday, May 23 to Friday, May 27 will explore the evolutionary dynamics of learning, especially as it applies to language and language evolution. “But there’ll be some people talking about biological evolution, a few maybe on cultural evolution,” Niyogi said. “The idea was to explore the ways in which learning and evolution manifest themselves in language, culture and biology.”

Learning theorists are attempting to understand the basic principles that describe how people or computers infer new knowledge from the facts at their disposal. Niyogi is among those scientists attempting to describe the dynamics of human learning in mathematical terms, just as scientists have created mathematical models for biological evolution.

“If you study the evolution of biological populations, the basic framework is Darwinian evolution and Mendelian inheritance,” he said. “Humans evolve, and the genes are transmitted from one generation to the next via inheritance.”

Evolutionary theorists have produced mathematical systems that show how gene sequences change from generation to generation and how biological diversity arises according to the laws of genetics. Learning theorists, meanwhile, look for the mathematical properties of evolutionary processes where the mode of transmission is learning rather than inheritance.

“A classic example of that is language evolution,” Niyogi said. “Languages are transmitted from parents to children, much like genes are transmitted, except that in the process of language you don’t really inherit the language of your parents in the way you inherit their genes.

“You actually have to learn their 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.”

The fields of learning theory and machine learning have different roots, but the Chicago events exemplify

some of their growing common interests. “The communities are mixing to a greater and greater extent,” Langford said.
Last modified at 01:56 PM CST on Monday, May 16, 2005.

University of Chicago News Office
5801 South Ellis Avenue - Room 200
Chicago, Illinois 60637-1473
(773) 702-8360
Fax: (773) 702-8324
Contact Us