Sakana AI Unveils PC-ALM, Training 1000-Layer Networks Without Backpropagation

Sakana AI introduced PC-ALM, a local learning method that trains 1000-layer neural networks without backpropagation, drawing commentary from LeCun. Researchers note, however, that the core idea dates back to 2018.

2026-09-15 ~ 2026-09-15 · 2 related posts

• Sakana AI's PC-ALM trains 1000-layer nets without backprop; researcher notes 2018 precedent — PMinervini · 2026-09-15
• Sakana's PC-ALM trains 1000-layer nets without backprop; LeCun calls it target prop reborn — ylecun · 2026-09-15