tower-of-babel-paradox-local-encoding-privacy
IN premise — summaries/2026/08/24/wiki-Manifold_hypothesis.md
Created 2026-08-24T17:11:18+00:00
The Tower of Babel Paradox (Brette, 2017) states that if each neuron's local encoding is maximally efficient (indistinguishable from random to neighboring neurons), then no global encoding or synchronization is possible across a large neural network, creating a theoretical obstruction to biologically plausible learning
Summary
If every neuron encodes its information in a way that looks random to its neighbors, the network as a whole can never coordinate or synchronize, which makes collective learning theoretically impossible. This sets a hard constraint on any brain-like system: you cannot simultaneously maximize local encoding privacy and maintain the global coordination that learning depends on, so any plausible model of neural learning must accept a tradeoff between the two.