Matthew Botvinick, a professor at Princeton who is now the head of neuroscience at DeepMind, Alphabet’s A.I. subsidiary, said that distilling relevant similarities and differences into vectors was “the secret sauce underlying all of these A.I. advances.”
What if learning/ education evolved away from storing data in memory as text, and towards a model closer to the way the brain works? The brain creates “vectors” of similarities and differences, new brain research has discovered, and uses those vectors to generate a component of what we call intelligence. It seems a time is coming when AI and human brains, will be working on tracks that converge into new forms of knowing and intelligence.
“Applying the new science of learning” is one of the 7 core elements PSA has promulgated for new educational effectiveness. That is no longer over the horizon, as advances in AI and brain science are proceeding at rapid rates.
What does that landscape look like today, and will education be nearly unrecognizable in the next five years? Here’s two articles that suggest new vistas we will need to grasp sooner than later to support new understandings of how to do learning and education.
DeepMind Says Its New AI Has Almost the Reading Comprehension of a High Schooler
The Science of Mind Reading | The New Yorker