Timezone: »

Baby Intuitions Benchmark (BIB): Discerning the goals, preferences, and actions of others
Kanishk Gandhi · Gala Stojnic · Brenden Lake · Moira R. Dillon

Tue Dec 07 04:30 PM -- 06:00 PM (PST) @

To achieve human-like common sense about everyday life, machine learning systems must understand and reason about the goals, preferences, and actions of other agents in the environment. By the end of their first year of life, human infants intuitively achieve such common sense, and these cognitive achievements lay the foundation for humans' rich and complex understanding of the mental states of others. Can machines achieve generalizable, commonsense reasoning about other agents like human infants? The Baby Intuitions Benchmark (BIB) challenges machines to predict the plausibility of an agent's behavior based on the underlying causes of its actions. Because BIB's content and paradigm are adopted from developmental cognitive science, BIB allows for direct comparison between human and machine performance. Nevertheless, recently proposed, deep-learning-based agency reasoning models fail to show infant-like reasoning, leaving BIB an open challenge.

Author Information

Kanishk Gandhi (Stanford University)
Gala Stojnic (New York University)
Brenden Lake (New York University)
Moira R. Dillon (New York University)

How do we build artificial intelligence that looks like human intelligence? I study infants’ and children’s precocious knowledge about the world, from the objects and places in it to the people and animals that act on those objects and move through those places. I ask how that knowledge derives from our evolutionary inheritance and how it forms the basis of uniquely human cultural and intellectual achievements. In doing so, I use my expertise in human commonsense to improve machine commonsense, creating AI that will better understand us and that we can better understand.

More from the Same Authors