Lassonde researchers win AI challenge with deep learning discovery

Though computers are getting better and better at meeting human needs, they remain poor communicators. Lassonde researchers are working on ways to improve AI comprehension. Professor Hui Jiang along with visiting researcher Quan Liu won first place at this year’s CommonSense Reasoning Competition held this July in New York City. In the Winograd Schema Challenge the team used deep learning to train a computer to recognize the relationship between different events, such as “playing basketball” and “winning” or “getting injured,” from thousands of texts. The Winograd Schema Challenge was proposed in 2014 as an improvement on the Turing Test.  The problem with the Turing Test is that it’s often easy for a program to fool a person using simple tricks and evasions. But a program cannot parse Winograd Schema or other ambiguous sentences without some…


Link to Full Article: Lassonde researchers win AI challenge with deep learning discovery