factual
Brains Converge Faster By Discarding Data
A human nervous system could not learn from 800 million image-caption pairs as CLIP-style models do; for us the world is learnable only because adjacent frames are related and we discard most information, taking in only what makes us more coherent, finding patterns as early as possible — yielding much faster convergence on far less data than statistical models that tolerate incoherent data.
factualpending
Speaker
Joscha BachEvidence Quote
“For us, the world is only learnable because the adjacent frames are related and we can afford to discard most of that information during learning... Our mind seems to be focused on finding the patterns as early as possible... It leads to much faster convergence. We only need to look the tiny fraction of the data to become coherent.”
Created: 6/13/2026, 3:31:54 AM
My Notes
Loading notes...