Memristive Networks Learn by Reorganizing: When the Material Is the Model
bravo_abad · x · 2026-09-20
In a post on his AI-for-Science newsletter Discovery at Scale, researcher Jorge Bravo Abad argues for a radical alternative to conventional AI hardware: instead of running a neural-network algorithm more efficiently on the material, let self-organizing memristive networks use the material's own dynamics as the computation itself.
These systems are random networks of nanowires or nanoparticles that learn by physically reorganizing themselves under electrical stimulation, embodying the model directly in hardware. The author frames this within his broader theme of making scientific discovery more systematic, scalable, and repeatable.
Related event: Memristive Networks Turn the Material Itself into the Model(2 posts)→
More from Research
- Open-weight science model Hydrogen-1.0 beats GPT-6 Astra on FSR at 72% lower cost — AKAMA199 · 2026-11-04
- Anthropic says it may imminently hit RSP threshold for AI-accelerated R&D, releases provable inference work — tobyordoxford · 2026-11-04
- Google scanned 1,033 real household products, packed into a 345 MB GLB for three.js — victormustar · 2026-11-04
- These 10 GPU optimization papers are why your model runs fast today — NandoDF · 2026-11-04
- INT8 isn't the problem: I benchmarked samplers on INT8 flow models — YakovSava · 2026-11-04
- Karpathy Boosts Open Hacker News Dataset: 10M+ Posts Usable as a Personal Knowledge Base — huangyun_122 · 2026-11-04