Rich Interaction: Teaching your Computer 'Why'

By August 21, 2009

Oregon State University researchers are developing a way to make computers learn why you want them to perform or not perform an action. This nuanced user-computer relationship -- rich interaction -- will make computers “a partner,” the developers say. "In the future we believe the computer should be like your partner," said Margaret Burnett, associate professor of computer science at Oregon State. "You help teach it, it gets to know you, you learn from each other, and it becomes more useful." In essence, what rich interaction does is add a ‘why’ component to computing. Explaining to your computer why you want a task done in a certain way helps it better learn your preferences.

"There are limits to what the computer can do just by its own observations and efforts to learn from experiences," Burnett said. "It needs to understand not just what it did right or wrong, but why. And for that, it has to continue interacting with human beings and make constant changes in its own programming, based on their feedback."

The developers also intend to simplify interactions, focusing on the average computer user. By adding the user’s dialogue to the computer’s own logic, the developers hope to create a more nuanced performance.

“[W]e want to develop algorithms that will allow the end user to ask the computer why it did something, read its response, and then explain why that was a mistake,” said Weng-Keen Wong, assistant professor of computer science at Oregon State.

This ability to recognize why something was a mistake is the next step towards real machine learning.

"For machine learning to reach its potential the computer and the user have to interact with each other in a fairly meaningful way, the computer really needs to get to know your situations and understand why it made a mistake, so that it can try not to make the same mistake again," Burnett said.

Legal mentions © L’Atelier BNP Paribas