Self-Taught: Software That Learns By Doing
Machine-learning techniques to create self-improving software are hitting the mainstream.
February 6, 2006 12:00 PM ETComputerworld -
Attempts to create self-improving software date to the 1960s. But "machine learning," as it's often called, has remained mostly the province of academic researchers, with only a few niche applications in the commercial world, such as speech recognition and credit card fraud detection. Now, researchers say, better algorithms, more powerful computers and a few clever tricks will move it further into the mainstream.

![]()
Stanford professor Sebastian Thrun with "Stanley," the car that used machine-learning techniques to drive itself 132 miles across the desert. ![]()
Thrun used several new machine-learning techniques in software that literally drove an autonomous car 132 miles across the desert to win a $2 million prize for Stanford in a recent contest put on by the Defense Advanced Research Projects Agency. The car learned road-surface characteristics as it went. And machine-learning techniques gave his team a productivity boost as well, Thrun says. "I could develop code in a day that would have taken me half a month to develop by hand," he says.
Computer scientist Tom Mitchell, director of the Center for Automated Learning and Discovery at Carnegie Mellon University, says machine learning is useful for the kinds of tasks that humans do easily -- speech and image recognition, for example -- but that they have trouble explaining explicitly in software rules. In machine-learning applications, software is "trained" on test cases devised and labeled by humans, scored so it knows what it got right and wrong, and then sent out to solve real-world cases.
Mitchell is testing the concept of having two classes of learning algorithms in essence train each other, so that together they can do better than either would alone. For example, one search algorithm classifies a Web page by considering the words on it. A second one looks at the words on the hyperlinks that point to the page. The two share clues about a page and express their confidence in their assessments.
Mitchell's experiments have shown that such "co-training" can reduce errors by more than a factor
Development
Additional Resources



White Papers & Webcasts
HP Technology Guide for Scalable Business Solutions
Download This Resource Now!
Enterprise Application Delivery: No User Left Behind
Gain the ability to deliver applications to all users, using any device, across any network.
Gartner: Magic Quadrant for Application Delivery Controllers, 2009
The market for products to improve the delivery of application software over networks remains dynamic and innovative. Vendors focused on solving enterprises' most-pressing...
Data Protection is not an insurance policy -you cannot buy-back lost data
Find out why you need to maintain access to critical information to run your business and remain competitive.
Chiquita selects Workday's fresh approach to Human Capital Management
A fresh approach to meet IT and HR objectives.
ITIL in Tough Economic Times
Are you looking for new inspiration to move forward with ITIL in these tough economic times?
The ROI of Software-As-A-Service
A Total Economic Impact™ Analysis Uncovers Long-Term Value In SaaS
IT Governance Podcast: IT Provider Forecasts $10 Million in Savings
In this podcast, learn how OTS was able to prioritize, then deliver, on the mission-critical demands and, in the process, project $10 million...
