Randomly connected units can self-assemble gradient-like learning, PNAS study shows
Dorian182 · reddit · 2026-10-15
MIT coverage (Oct 13) highlights a PNAS paper published Oct 9, "How Biological Synapses Self-Assemble Gradient Learning." Key points:
- Randomly interconnected neuron-like units governed by shared plasticity rules can self-assemble learning motifs and perform pattern recognition and image classification, matching the performance of backpropagation-trained networks in reported comparisons.
- The paradigm shift: instead of designing a learning circuit directly, you specify local rules from which the learning machinery organizes itself.
- The authors caution this doesn't show the brain uses this mechanism — the motivation is biological plausibility, not a replacement for backprop at scale.
The poster notes stability of motifs across initial connectivity and generalization beyond demonstrated tasks are key questions for evaluating the result.
Related event: MIT Study Shows Neural Networks Can Self-Assemble Learning Mechanisms(3 posts)→
More from Research
- RExBench: best coding agent completes only 33% of autonomous research extensions — najoungkim · 2026-10-16
- Google's CLASHarness benchmarks multi-agent conflicts in computer use, finds frontier VLMs struggle — google · 2026-10-16
- Tsinghua team proves vanilla flow-matching sampling can't exceed constant-depth expressive power — SQZ · 2026-10-16
- APEX-Harm benchmark: script-based injection attacks easily compromise agents in shared workspaces — KAUST · 2026-10-16
- Residual-stream forecasts predict agent violations before they happen, USC study shows — USC · 2026-10-16
- KAUST's NoteBank adds sparse episodic memory to linear attention, boosting long-context for free — kaust-generative-ai · 2026-10-16