Pilot corpus: NIH-funded biomedical datasets. Any DOI can be scored on demand — network effects sharpen as coverage grows.

The training process of many deep networks explores the same low-dimensional manifold

Proceedings of the National Academy of Sciences(2024)10.1073/pnas.2310002121Source: DataRank Database

The training process of many deep networks explores the same low-dimensional manifold is a research paper published in Proceedings of the National Academy of Sciences (2024). On theSindex it has a DataRank of 0. It has been cited 11 times.

N/A

0DataRank · unranked

Scored on demand from live citation data

Open Access11 citations

Download PDF

Cite:

Linked data & code

The Training Process of Many Deep Networks Explores the Same Low-Dimensional Manifold · arXiv

DataRank reads this dataset's downstream impact straight off the citation graph — no black box, no proprietary weighting. How is this computed?

›Methodology & internals

datarank_citation_only_1hop_v6· scope data_onlyMethodology

Pipeline:MetadataData-paper checkEnrichmentCitation networkScoring

Enrichment:Funding (14 grants)Impact (FWCI 2.25)OA: hybridTopics (38)Linked Datasets (1)DataCite Reuse (1)IDs (PubMed, PMC, OpenAlex)

FAIR Checklist

Context only (not used in score)

Findable (2/2)

Has DOI
Indexed in repositories

Accessible (1/2)

Open Access

Interoperable (2/2)

DataCite relations
Linked datasets

Reusable (0/3)

FAIR checklist signals are shown for context only and do not affect DataRank scoring.

Run a calibrated FAIR evaluation for this paper →

We only score data papers we can read in full — never from an abstract alone.

Authors (8)

Itay Griniasty ORCID,Han Kheng Teoh ORCID,Rahul Ramesh,Rubing Yang,Mark K. Transtrum ORCID

Live enrichment

PubMed:38470929 PMC:PMC10962999 OpenAlex:https://openalex.org/W4392696399

National Science Foundation

Grant: IIS-2145164

National Science Foundation

Grant: CCF-2212519

DOD | USN | Office of Naval Research

Grant: N00014-22-1-2255

National Science Foundation

Grant: DMREF-89228

National Science Foundation

Grant: EFRI-1935252

HHS | National Institutes of Health

Grant: 1R01NS116595-01

Determining computational principles governing neural circuits responsible for feedback and movement control of D. melanogaster flight

National Science Foundation

Grant: DMR-1719490

National Science Foundation

Grant: DMR- 1753357

NINDS NIH HHS

Grant: R01 NS116595

National Science Foundation

Grant: 1935252

EFRI C3 SoRo: Micron-scale Morphing Soft-Robots for Interfacing With Biological Systems

National Science Foundation

Grant: 1719490

Exploiting emergent scale invariance

National Science Foundation

Grant: 1753357

CAREER: Connecting Mathematical Models Across Scales

National Science Foundation

Grant: 2145164

CAREER: Foundations of Small Data

National Science Foundation

Grant: 2212519

Collaborative Research: RI: Medium: MoDL: Occams Razor in Deep and Physical Learning

FWCI

2.25

Citation Percentile

0.9%

Influential Citations

Citation Trend

2023

2024

2025

2026

hybridLicense: cc-by-nc-nd

journalhttps://doi.org/10.1073/pnas.2310002121

HYBRIDhttps://doi.org/10.1073/pnas.2310002121

publisherhttps://doi.org/10.1073/pnas.2310002121

publisherhttps://pnas.org/doi/pdf/10.1073/pnas.2310002121

repositoryhttps://pubmed.ncbi.nlm.nih.gov/38470929

Fields of Study

Neural Networks and ApplicationsGenerative Adversarial Networks and Image SynthesisFace and Expression RecognitionMedicineComputer SciencePhysicsMathematics0301 basic medicine0303 health sciences03 medical and health sciences

Keywords

Manifold (fluid mechanics)Nonlinear dimensionality reductionManifold alignmentComputer scienceArtificial intelligenceRegularization (linguistics)Range (aeronautics)Process (computing)Deep learningSpace (punctuation)Probabilistic logicMachine learningDimensionality reductionEngineeringOptimizationPrincipal component analysisVisualizationInformation GeometryFOS: Computer and information sciencesComputer Science - Machine LearningPhysical SciencesFOS: Physical sciencesDisordered Systems and Neural Networks (cond-mat.dis-nn)Differential geometric aspects of statistical manifolds and information geometryCondensed Matter - Disordered Systems and Neural NetworksArtificial neural networks and deep learningInformation geometry (statistical aspects)Machine Learning (cs.LG)

The Training Process of Many Deep Networks Explores the Same Low-Dimensional Manifold

10.48550/arxiv.2305.01604arXiv