# Sakana AI — Training 1000-layer networks without backpropagation

- Company: Sakana AI (sakana.ai)
- Announced: 2026-09-13T15:00:00+00:00
- Category: research-paper
- Coverage: not counted
- Announcement: yes
- Group: announcements
- Source: https://sakana.ai/pc-alm
- Record: https://forck.live/items/14458-training-1000-layer-networks-without-backpropagation
- Subject: Sakana models

Sakana AI introduced PC-ALM, a local-learning alternative to backpropagation that trains 1000-layer neural networks using only local dynamics. The method generalizes predictive coding with an augmented Lagrangian, introducing dual neurons (Lagrange multipliers) as part of layer-local dynamics, enabling signal propagation to arbitrary depth in deep narrow networks where standard predictive coding struggles.

## Evidence

Verbatim from https://sakana.ai/pc-alm:

> Our method trains 1000-layer neural nets using only local dynamics, and without backprop.

---

Record: https://forck.live/items/14458-training-1000-layer-networks-without-backpropagation
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md