Though computers are getting better and better at meeting human needs, they remain poor communicators.
Hui Jiang
Lassonde School of Engineering researchers are working on ways to improve artificial intelligence (AI) comprehension, and Professor Hui Jiang, along with researcher Quan Liu, won first place at this year's CommonSense Reasoning Competition held in July in New York City.
Jiang is a professor in the Department of Electrical Engineering & Computer Science and Liu is a visiting student from the University of Science & Technology of China.
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.
According to New York University, a Winograd schema is “a pair of sentences that differ in only one or two words and that contain an ambiguity that is resolved in opposite ways in the two sentences and requires the use of world knowledge and reasoning for its resolution.”
The Winograd Schema Challenge was proposed in 2014 as an improvement on the Turing Test, which was found to provide less authentic results due to programs being able to fool a person using simple tricks and evasions. However, it was found that a program cannot parse a Winograd Schema or other ambiguous sentences without some form of general knowledge.
The challenge had four contestants with the Lassonde group achieving the highest score of 58 per cent correct.
The cutting-edge work by the Lassonde team promises to create more intuitive AI interfaces in the future.
"Artificial intelligence is one of the most exciting fields of research currently, and it's thrilling to see Lassonde researchers doing work that's going to transform the way we interact with machines," said Lassonde's founding Dean Janusz Kozinski.