AI has a problem with generalization. Human brains may offer a solution.

a b a a a b a a large
Image: Ramachandra Babu/Gulf News

Unconstrained by evolution, AI has the potential to churn through vast amounts of data to surpass our puny, fatty central processors. Yet even as a single algorithm beats humans in a specific problem—chess, Go, Dota, medical diagnosis for breast cancer—humans kick their butts every time when it comes to a brand new task. Somehow, the innate structure of our brains, when combined with a little worldly experience, lets us easily generalize one solution to the next. State-of-the-art deep learning networks can’t.

In a new paper published in Neuron, [Dr. Andreas] Tolias and colleagues in Germany argue that more data or more layers in artificial neural networks isn’t the answer. Rather, the key is to introduce inductive biases—somewhat analogous to an evolutionary drive—that nudge algorithms towards the ability to generalize across drastic changes.

Related article:  'Wild and frightening possibilities': How AI and genomics could affect reproduction

“The next generation of intelligent algorithms will not be achieved by following the current strategy of making networks larger or deeper. Perhaps counter-intuitively, it might be the exact opposite…we need to add more bias to the class of [AI] models,” the authors said.

What better source of inspiration than our own brains?


Read full, original post: Deep Learning Networks Can’t Generalize—But They’re Learning From the Brain

Outbreak Daily Digest
Biotech Facts & Fallacies
GLP Podcasts
Infographic: Here’s where GM crops are grown around the world today

Infographic: Here’s where GM crops are grown around the world today

Do you know where biotech crops are grown in the world? This updated ISAAA infographics show where biotech crops were ...
News on human & agricultural genetics and biotechnology delivered to your inbox.
glp menu logo outlined

Newsletter Subscription

* indicates required
Email Lists
Send this to a friend