The researchers argue that their system mimics the human ability to learn new concepts from few examples. They say it offers hope that the type of computational structure it’s built on, called a probabilistic program, could help model human acquisition of more sophisticated concepts as well.
“In the current AI landscape, there’s been a lot of focus on classifying patterns,” said Josh Tenenbaum, a professor in the Department of Brain and Cognitive sciences at MIT. “But what’s been lost is that intelligence isn’t just about classifying or recognising; it’s about thinking. This is partly why, even though we’re studying hand-written characters, we’re not shy about using a word like ‘concept.’”
The researchers subjected their system to a battery of tests. In one, they presented it with a single example of a character in a writing system it had never seen before and asked it to produce nine variations on the same character. In another test, they presented it with several characters in an unfamiliar writing system and asked it to produce new characters that were in some way similar. And in a final test, they asked it to make up entirely new characters in a hypothetical writing system.
Human subjects were then asked to perform the same three tasks. Finally, a separate group of human judges was asked to distinguish the human subjects’ work from the machine’s. Across all three tasks, the judges could identify the machine outputs with about 50% accuracy — no better than chance.
Conventional machine-learning systems — such as the ones that led to the speech-recognition algorithms on smartphones — often perform very well on constrained classification tasks, but they must first be trained on huge sets of training data. Humans, by contrast, frequently grasp concepts after just a few examples. That type of ‘one-shot learning’ is something that the researchers designed their system to emulate.
Like a human subject, however, the system comes to a new task with substantial background knowledge, which in this case is captured by a probabilistic program. Whereas a conventional computer program systematically decomposes a high-level task into its most basic computations, a probabilistic program requires only a vague model of the data it will operate on.
Armed with that model, the system then analysed hundreds of motion-capture recordings of humans drawing characters in several different writing systems, learning statistics on the relationships between consecutive strokes and substrokes as well as on the variation tolerated in the execution of a single stroke.
