Scientists have made a breakthrough in creating a pocket-sized AI brain, thanks to the help of monkey neurons. Researchers were able to shrink an AI vision model to a tiny fraction of its original size, using data from macaque monkeys. This achievement hints at how living brains can do so much with so little power, and it could have significant implications for the future of artificial intelligence.
The AI model, which mimics a part of the brain's visual system, started with 60 million variables. However, the team was able to compress it into a version that performed nearly as well using just 10,000 variables. This is an incredibly small model, according to Ben Cowley, an author of the study and an assistant professor at Cold Spring Harbor Laboratory. He notes that this model is something that could be sent in a tweet or an email.
The compact model also appears to work more like a living brain, which could help scientists study what goes wrong in diseases like Alzheimer's. By understanding how the AI model replicates strategies found in nature, scientists may be able to gain insights into the inner workings of the human brain, says Mitya Chklovskii, a group leader at the Simons Foundation's Flatiron Institute.
The study is part of an effort to understand the human visual system, which takes in bits of light and transforms them into something we recognize, like grandma or the Grand Canyon. Cowley says scientists who study the visual system have been trying to answer questions like, 'How do you recognize a cat?' or 'How do you recognize a dog?'
However, there's a problem: 'We're very impoverished in our understanding of how these AI systems work,' Cowley says, 'much like our own brain.' To address this, Cowley and his team created an AI model that they could understand. It simulates just one part of the visual system, which features cells called V4 neurons.
These neurons encode colors, textures, curves, and very complicated proto-objects. Existing AI systems can do the same thing using deep neural network models, which require powerful computers and learn by considering a huge range of possibilities. But Cowley's team was after something more efficient.
'We want to take these big clunky models and try to compress it down into a much smaller, compact form,' he says. They started with a model trained on data from macaque monkeys, then looked for parts of the model that were redundant or unnecessary. They also applied statistical techniques like those used to compress digital photos.
The result: a model small enough to put in an email attachment. Because the model is so small and simple, the team was able to get a glimpse of what its artificial neurons were doing. Some V4 neurons, for example, were responding to shapes with strong edges and lots of curves, while others seemed to respond only to small dots in an image.
The specialized nature of these V4 neurons may help explain how human and other primate brains are able to make sense of what they see without relying on massive computing power. The findings also may have implications for artificial intelligence. If our brains have less complex models and yet can do more than these AI systems, that tells us something about our AI systems, Cowley says. Namely, they could probably be smaller and simpler yet still do a better job interpreting what they see.
For example, self-driving cars might be able to run on less powerful computers, he says, while correctly distinguishing a pedestrian from an airborne plastic bag. However, AI systems need to do more than shrink in order to perform as well as a human brain, Chklovskii says. For example, a person can easily recognize a friend's face in any setting and from many angles, even if that friend has acquired a suntan or is sporting a new haircut. AI systems struggle with this sort of task, even when powered by supercomputers.
That may be because current AI models are based on an understanding of the human brain from the 20th century, Chklovskii says. 'Since then, we learned a lot more about the brain,' he says. 'So maybe we should update the foundations of the artificial networks.'
