Data Science Investment Counter —Funding raised by UK data science companies in 2018.
£ 5.640 Million

Researchers create AI that can play Pictionary

Scientists from the Allen Institute for Artificial Intelligence (AI2) have released an online Pictionary-style game called Iconary yesterday, which they say could help sharpen AI’s common sense.

In Iconary, players must try and draw complex scenes like “a crowd celebrating a victory” or “kicking a tennis ball on a beach” while a computer program named AllenAI tries to guess what they’ve drawn. The focal point of the research lies in the fact that the AI has never seen these phrases before.

“Iconary is one of the first times an AI system is paired in a collaborative game with a human player instead of antagonistically working against them,” the Iconary website states.

“AllenAI is capable of both understanding and producing a nearly infinite combination of real-world scenarios represented in the phrases in Iconary, a compelling example of the potential of common sense for AI and the power of human-AI collaboration.”

Understanding these combinations is challenging for AI, says AI2 researcher Ani Kembhavi, because it tests a wide range of common sense skills. The algorithms must begin by identifying the visual elements in the picture, figure out how they relate to one another, then translate that scene into simple language that humans can understand.

Researchers said this AI, because of its news skills, could teach computers information that other AI benchmarks like Go and StarCraft can’t.

“This is a first step toward exploiting common sense,” said Aniruddha Kembhavi, a computer scientist who worked on the project.

Fellow computer scientist Ali Farhadi added: “We wanted to build an AI system that can collaborate with human beings, and at the same time is learning about how humans think, how they act… I actually kind of feel that this system is connecting to me deep in my thoughts.

“This is a whole different way of thinking about AI and games,” Farhadi said, “and what you can actually learn from playing a game other than just being the best at playing that game,” he said. “Because of that, I believe we are going to spend a fair amount of time on this. I certainly hope that looking back at this in five years, this will really be … an icon in that space.”

Featured Image via Jeremy Keith/Flickr

Co-working space and blog dedicated to all things data science.

Subscribe to our newsletter