Self-Taught: Software That Learns By Doing

Machine-learning techniques to create self-improving software are hitting the mainstream.

By Gary Anthes

February 6, 2006 12:00 PM ET

Recommended

Computerworld - 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.

And as the technology grows, so does the need for it. "In the past, someone would look at a problem, write some code, test it, improve it by hand, test it again and so on," says Sebastian Thrun, a computer science professor at Stanford University and the director of the Stanford Artificial Intelligence Laboratory. "The problem is, software is becoming larger and larger and less and less manageable. So there's a trend to make software that can adapt itself. This is a really big item for the future."
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

Recommend Digg Twitter ShareThis

Email Print Send Feedback Reprints

Jump to comments

Development

Additional Resources

Xerox

Win a color printer!

By using solid ink technology only from Xerox, you could save up to 65% by printing color for the cost of black and white. Enter for a chance to WIN a PhaserTM 8860 network color printer!

Microsoft

Upgrade to Internet Explorer 8 today.

Save time and mitigate security risk. Deploy it now.

Sybase

NEXT-GENERATION MOBILITY PLATFORMS

In this white paper, IDC analyzes the role of next-generation mobile enterprise platforms as organizations seek a more strategic deployment of mobile solutions.

Learn the important issues you must consider before starting your next mobility initiative. Get your mobility white paper from IDC now, compliments of Sybase.

White Papers & Webcasts

Informatica 9 Launch: Transform your Business. Transform your world.

LIVE Nov 10, 2009 12:00 PM ET

Architecting Business Intelligence Applications for Change: The Open Solution

LIVE Nov 12, 2009 02:00 PM ET

Faster, Cheaper and Easier to Maintain
Can you afford not to upgrade your servers to today's advanced, energy-efficient technologies?

Top to Bottom Performance Management Excellence at the City of Chicago
View now.

Do more with less thru Netcool?
Learn how IBM Tivoli® Netcool® solutions can help service providers streamline their operations, improve responsiveness and reduce costs.

Effectively Implementing Datacenter Automation
Effectively select and deploy the best datacenter automation solution today!

IDC report: Profitability and OSS Support: A Return on Investment Analysis of IBM Tivoli Netcool
IDC studied 14 mobile and fixed-line service providers that implemented Tivoli® Netcool® and found that IBM Tivoli Netcool can help in big ways.

Aligning IT to Business: The Rising Importance of Application Delivery Networks
Application Delivery Networking (ADN) will play a vital role in helping enterprises incorporate strategic technologies to achieve business initiatives.

IBM Systems Makeover Analysis for Oracle Environments
This brochure shows how the IBM Systems Makeover Analysis takes a look at your current Oracle hardware infrastructure, then proposes a high-level future...

Strategies for Driving Down IT Support Costs with Lite ITIL Practices
View this now!

Lower your IT costs and risks: Get a server makeover
Find out how a server makeover analysis can help you develop a high-level roadmap for your infrastructure.

Mitigate Risk, Lower Costs and Improve Network Efficiency
Create a stable IP network that not only meets today's challenges, but is flexible enough to also meet future demands.

All White Papers | All Webcasts