dorsal/arxiv
View SchemaNeural Decision Boundaries for Maximal Information Transmission
| Authors | Tatyana Sharpee, William Bialek |
|---|---|
| Categories | |
| ArXiv ID | q-bio/0703046 |
| URL | https://arxiv.org/abs/q-bio/0703046 |
| DOI | 10.1371/journal.pone.0000646 |
Abstract
We consider here how to separate multidimensional signals into two categories, such that the binary decision transmits the maximum possible information transmitted about those signals. Our motivation comes from the nervous system, where neurons process multidimensional signals into a binary sequence of responses (spikes). In a small noise limit, we derive a general equation for the decision boundary that locally relates its curvature to the probability distribution of inputs. We show that for Gaussian inputs the optimal boundaries are planar, but for non-Gaussian inputs the curvature is nonzero. As an example, we consider exponentially distributed inputs, which are known to approximate a variety of signals from natural environment.
{
"annotation_id": "56b64576-82c8-4c4a-98a9-5be936407299",
"date_created": "2026-03-02T18:01:35.432000Z",
"date_modified": "2026-03-02T18:01:35.432000Z",
"file_hash": "2fa2250e2ae9fdadd6f8dd8ef20e1b9c35e0620553efbf6f175fb0f5ef246bf4",
"private": false,
"record": {
"abstract": "We consider here how to separate multidimensional signals into two\ncategories, such that the binary decision transmits the maximum possible\ninformation transmitted about those signals. Our motivation comes from the\nnervous system, where neurons process multidimensional signals into a binary\nsequence of responses (spikes). In a small noise limit, we derive a general\nequation for the decision boundary that locally relates its curvature to the\nprobability distribution of inputs. We show that for Gaussian inputs the\noptimal boundaries are planar, but for non-Gaussian inputs the curvature is\nnonzero. As an example, we consider exponentially distributed inputs, which are\nknown to approximate a variety of signals from natural environment.",
"arxiv_id": "q-bio/0703046",
"authors": [
"Tatyana Sharpee",
"William Bialek"
],
"categories": [
"q-bio.NC"
],
"doi": "10.1371/journal.pone.0000646",
"title": "Neural Decision Boundaries for Maximal Information Transmission",
"url": "https://arxiv.org/abs/q-bio/0703046"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "8c996064-6660-4a20-b0d4-1fb30a7cd8ed",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}