dorsal/arxiv
View SchemaA Global Algorithm for Training Multilayer Neural Networks
| Authors | Hong Zhao, Tao Jin |
|---|---|
| Categories | |
| ArXiv ID | physics/0607046 |
| URL | https://arxiv.org/abs/physics/0607046 |
Abstract
We present a global algorithm for training multilayer neural networks in this Letter. The algorithm is focused on controlling the local fields of neurons induced by the input of samples by random adaptations of the synaptic weights. Unlike the backpropagation algorithm, the networks may have discrete-state weights, and may apply either differentiable or nondifferentiable neural transfer functions. A two-layer network is trained as an example to separate a linearly inseparable set of samples into two categories, and its powerful generalization capacity is emphasized. The extension to more general cases is straightforward.
{
"annotation_id": "e736c7ee-5de3-4955-9a25-0e0dd9bdd75d",
"date_created": "2026-03-02T18:01:11.094000Z",
"date_modified": "2026-03-02T18:01:11.094000Z",
"file_hash": "33862bab476aff048875e85a93060dc1bd1a9250380d6ffe6baec40ee06fe083",
"private": false,
"record": {
"abstract": "We present a global algorithm for training multilayer neural networks in this\nLetter. The algorithm is focused on controlling the local fields of neurons\ninduced by the input of samples by random adaptations of the synaptic weights.\nUnlike the backpropagation algorithm, the networks may have discrete-state\nweights, and may apply either differentiable or nondifferentiable neural\ntransfer functions. A two-layer network is trained as an example to separate a\nlinearly inseparable set of samples into two categories, and its powerful\ngeneralization capacity is emphasized. The extension to more general cases is\nstraightforward.",
"arxiv_id": "physics/0607046",
"authors": [
"Hong Zhao",
"Tao Jin"
],
"categories": [
"physics.bio-ph",
"physics.comp-ph"
],
"title": "A Global Algorithm for Training Multilayer Neural Networks",
"url": "https://arxiv.org/abs/physics/0607046"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "f771bf01-be61-46a9-a932-7411a4e3e289",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}