dorsal/arxiv
View SchemaForward-only learning in memristor arrays with month-scale stability
| Authors | Adrien Renaudineau, Mamadou Hawa Diallo, Théo Dupuis, Bastien Imbert, Mohammed Akib Iftakher, Kamel-Eddine Harabi, Clément Turck, Tifenn Hirtzlin, Djohan Bonnet, Franck Melul, Jorge-Daniel Aguirre-Morales, Elisa Vianello, Marc Bocquet, Jean-Michel Portal, Damien Querlioz |
|---|---|
| Categories | |
| ArXiv ID | 2601.09903vv1 |
| URL | https://arxiv.org/abs/2601.09903 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
Turning memristor arrays from efficient inference engines into systems capable of on-chip learning has proved difficult. Weight updates have a high energy cost and cause device wear, analog states drift, and backpropagation requires a backward pass with reversed signal flow. Here we experimentally demonstrate learning on standard filamentary HfOx/Ti arrays that addresses these challenges with two design choices. First, we realize that standard filamentary HfOx/Ti memristors support sub-1 V reset-only pulses that cut energy, improve endurance, and yield stable analog states. Second, we rely on forward-only training algorithms derived from Hinton's Forward-Forward that use only inference-style operations. We train two-layer classifiers on an ImageNet-resolution four-class task using arrays up to 8,064 devices. Two forward-only variants, the double-pass supervised Forward-Forward and a single-pass competitive rule, achieve test accuracies of 89.5% and 89.6%, respectively; a reference experiment using backpropagation reaches 90.0%. Across five independent runs per method, these accuracies match within statistical uncertainty. Trained models retain accuracy for at least one month under ambient conditions, consistent with the stability of reset-only states. Sub-1 V reset updates use 460 times less energy than conventional program-and-verify programming and require just 46% more energy than inference-only operation. Together, these results establish forward-only, sub-1 V learning on standard filamentary stacks at array scale, outlining a practical, pulse-aware route to adaptive edge intelligence.
{
"annotation_id": "96e8ef31-5a41-4061-b210-70e62a301cd4",
"date_created": "2026-02-17T05:53:24.376000Z",
"date_modified": "2026-02-17T05:53:24.376000Z",
"file_hash": "c7af2a45eaa4a59817b4cdbd21633b4f9cbc28f9899bdccbc3dd7f3e4d63f79f",
"private": false,
"record": {
"abstract": "Turning memristor arrays from efficient inference engines into systems capable of on-chip learning has proved difficult. Weight updates have a high energy cost and cause device wear, analog states drift, and backpropagation requires a backward pass with reversed signal flow. Here we experimentally demonstrate learning on standard filamentary HfOx/Ti arrays that addresses these challenges with two design choices. First, we realize that standard filamentary HfOx/Ti memristors support sub-1 V reset-only pulses that cut energy, improve endurance, and yield stable analog states. Second, we rely on forward-only training algorithms derived from Hinton\u0027s Forward-Forward that use only inference-style operations. We train two-layer classifiers on an ImageNet-resolution four-class task using arrays up to 8,064 devices. Two forward-only variants, the double-pass supervised Forward-Forward and a single-pass competitive rule, achieve test accuracies of 89.5% and 89.6%, respectively; a reference experiment using backpropagation reaches 90.0%. Across five independent runs per method, these accuracies match within statistical uncertainty. Trained models retain accuracy for at least one month under ambient conditions, consistent with the stability of reset-only states. Sub-1 V reset updates use 460 times less energy than conventional program-and-verify programming and require just 46% more energy than inference-only operation. Together, these results establish forward-only, sub-1 V learning on standard filamentary stacks at array scale, outlining a practical, pulse-aware route to adaptive edge intelligence.",
"arxiv_id": "2601.09903",
"authors": [
"Adrien Renaudineau",
"Mamadou Hawa Diallo",
"Th\u00e9o Dupuis",
"Bastien Imbert",
"Mohammed Akib Iftakher",
"Kamel-Eddine Harabi",
"Cl\u00e9ment Turck",
"Tifenn Hirtzlin",
"Djohan Bonnet",
"Franck Melul",
"Jorge-Daniel Aguirre-Morales",
"Elisa Vianello",
"Marc Bocquet",
"Jean-Michel Portal",
"Damien Querlioz"
],
"categories": [
"cs.ET"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "Forward-only learning in memristor arrays with month-scale stability",
"url": "https://arxiv.org/abs/2601.09903",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "0695a754-f753-4b2f-abc2-3dfc2f6c8fef",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}