MIT: neural nets self-assemble learning circuits with local rules, no backprop
MacrinePhD · x · 2026-10-25
New research from MIT's McGovern Institute shows artificial neural networks can build their own learning machinery using simple local rules (like "neurons that fire together, wire together") instead of engineered back-propagation.
- Using self-assembling motifs, randomly connected units spontaneously organized into functional circuits.
- These circuits achieved image classification and pattern recognition performance matching conventionally trained models.
- Published in PNAS, the work marks a step toward biologically plausible AI and understanding brain plasticity — the brain has no external designer, yet learning machinery emerges on its own.
More from Research
- Training data-poisoning victim speaks out; ByteDance's Xinggang Wang says VAR路线皆 bitter lesson — zibuyu9 · 2026-10-25
- JevBench v1.7 adds a 10-second rule: late decisions count as wrong in agent benchmark — airesearch12 · 2026-10-25
- Open-source virtual human body: 700 muscle actuators, 206 joints, 90 segments — mtizard · 2026-10-25
- EMNLP paper: users trust model verdicts over reasoning, confident tone hides errors — MaartenSap · 2026-10-25
- Prime Intellect merges IcePop score centering with truncated sampling replay in prime-rl — samsja19 · 2026-10-25
- EMBER Targets Token Embeddings: New Method Blocks LLM Knowledge Relearning — megamor2 · 2026-10-25