Children learn general representations with very few explicitly labeled examples, unlike supervised learning approaches like AlexNet that required the enormous manually-labeled ImageNet dataset, suggesting human learning relies on different mechanisms than supervised learning.
factualpending
Evidence Quote
“children are able to learn very general representations for concepts like dog with very few explicitly labeled examples”
Created: 8/12/2026, 6:35:53 PM
Your notes
Loading notes...