# Sakana AI trains 1,000-layer networks without backpropagation

The method accumulates per-layer constraint errors into a local Lagrange multiplier and lands within about two points of backpropagation accuracy on MNIST at that depth.

- Published: 2026-09-15T06:27:29.669Z
- Canonical: https://polylog.news/ai/2026-09-15/sakana-ai-trains-1-000-layer-networks-without-backpropagatio
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [Polylog editors](https://polylog.news)

Sakana AI has published Augmented Lagrangian Predictive Coding, a training method that replaces the global backward pass with updates computed only from interactions between neighboring layers. Predictive coding has long been proposed as a…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-09-15/sakana-ai-trains-1-000-layer-networks-without-backpropagatio (subscription information: https://polylog.news/pricing).