An INSIDER story this month highlighted efforts from USC researchers to train artificial intelligence to imagine the unseen. The artificial-intelligence system uses the attributes that it "knows" to then think up a never-before-seen object.
The approach takes a group of sample images — rather than one sample at a time as traditional algorithms have done — and mines the similarity between them.
The design technique, according to the project’s lead researcher, could lead to fairer A.I., new medicines, and increased autonomous vehicle safety.
“We can use the learned experience to synthesize or imagine some extreme or dangerous situation, which can teach the self-driving system to avoid this situation.” USC researcher Yunhao Ge told Tech Briefs.
What do you think? Does A.I. Need to Be Able to ‘Imagine?’
Share your questions and comments on our Question of the Week page.