dl-unsupervised-more-biologically-plausible-than-backprop

IN premise — entries/2026/06/21/wiki-Deep_learning-chunk-5.md

Created 2026-06-21T09:55:49+00:00

Unsupervised deep learning methods (generative models, deep belief networks) are considered more biologically plausible than supervised backpropagation.

Dependents

These beliefs depend on this one:

• IN biological-plausibility-inversely-correlated-with-pragmatic-adoption — Unsupervised deep learning methods are considered more biologically plausible than backpropagation, yet pragmatism's filtering of biological inspiration tends to retain efficiency properties (local connectivity, weight sharing, gating) while discarding robustness properties (redundancy, homeostasis, graceful degradation) — suggesting that the most biologically faithful approaches may be among the least pragmatically favored, and that the biological inspirations most readily adopted are those that enhance capability rather than reliability.