One of the challenges in developing artificial intelligence that can allow robots to effectively interact with people involves the capability to understand and respond to human behaviors. Researchers at Stanford University believe that they have discovered a way to address this issue, and that is by letting the A.I. read literature.
The A.I., which is called Augur, has been made to access the popular online writing community Wattpad with its hundreds of thousands of stories. The researchers expect that the massive information found in fiction will enable robots to be more effective in predicting human behavior and their responses to particular situations and actions. More details about this project can be accessed in this link.
"Over many millions of words, these mundane patterns [of people's reactions] are far more common than their dramatic counterparts," the researchers said, according to Engadget. "Characters in modern fiction turn on the lights after entering rooms; they react to compliments by blushing; they do not answer their phones when they are in meetings."
Teaching robots to read in order to understand humans and learn to adopt to their behaviors and actions is no longer a groundbreaking idea. There are several A.I. projects today that employ the same approach. For instance, Facebook's neural networks are being trained to understand more about human behavior by reading loads of children's books, New Scientist reported.