Sakana AI Paper: Neural Networks Can Learn While Respecting Biological Principles
TheTuringPost · x · 2026-07-31
Explores whether building AI requires exactly copying the brain's structure. The human brain follows Dale's principle, where each neuron is strictly either excitatory or inhibitory, a rule most artificial neural networks ignore.
A paper by Sakana AI investigated if neural networks can learn difficult tasks while strictly obeying this rule:
- Researchers separated excitatory and inhibitory signals and abandoned standard backpropagation.
- They used an Error Diffusion (ED) method that doesn't require transporting weights backward.
- The network achieved 96.7% accuracy on MNIST and 61.7% on CIFAR-10.
This shows biologically plausible learning is possible, though not yet matching conventional networks. Like how building airplanes didn't require mimicking bird wings, studying the brain can still offer valuable inspirations for AI.
Related event: Sakana AI Explores Biologically-Constrained Neural Networks(3 posts)→
More from Research
- LightOn Open-Sources 149M Param Retrieval Models Achieving SOTA — antoine_chaffin · 2026-07-31
- Princeton Lecture Explains Why Enormous Random Matrices Stop Behaving Randomly — tctjr · 2026-07-31
- New High Score: Team Achieves 67.5% on ARC-AGI-2 Benchmark — JFPuget · 2026-07-31
- Nature Neuroscience: Mice Curiosity Linked to Orbitofrontal Cortex Cells — KanakaRajanPhD · 2026-07-31
- Oopsie-Data Project Opens for Collaboration: Learning from Failures in RL — PeterStone_TX · 2026-07-31
- UCSB Researchers Awarded SIGGRAPH 2026 Test-of-Time Award — ninamiolane · 2026-07-31