I think it's more fair to say humans were "trained" over millions of years of transfer learning, and an individual human is fine tuned using much less data than Chinchilla.
I think humans and current deep learning models are running sufficiently different algorithms that the scaling curves of one don't apply to the other. This needn't be a huge difference. Convolutional nets are more data efficient than basic dense nets.
3Yitzi Litt
Is that fair to say? How much kolmogorov complexity can be encoded by evolution at a maximum, considering that all information transferred through evolution must be encoded in a single (stem) cell? Especially when we consider how genetically similar we are to beings which don’t even have brains, I have trouble imagining that the amount of “training data” encoded by evolution is very large.
I think it's more fair to say humans were "trained" over millions of years of transfer learning, and an individual human is fine tuned using much less data than Chinchilla.