Journal article
Deep neural network classifier for multidimensional functional data
Scandinavial Journal of Statistics, Vol.50(4), pp.1667-1686
12/2023
DOI: 10.1111/sjos.12660
Abstract
We propose a new approach, called as functional deep neural network (FDNN), for classifying multidimensional functional data. Specifically, a deep neural network is trained based on the principal components of the training data which shall be used to predict the class label of a future data function. Unlike the popular functional discriminant analysis approaches which only work for one-dimensional functional data, the proposed FDNN approach applies to general non-Gaussian multidimensional functional data. Moreover, when the log density ratio possesses a locally connected functional modular structure, we show that FDNN achieves minimax optimality. The superiority of our approach is demonstrated through both simulated and real-world datasets.
Details
- Title: Subtitle
- Deep neural network classifier for multidimensional functional data
- Creators
- Shuoyang Wang - Yale UniversityGuanqun Cao - Auburn UniversityZuofeng Shang - New Jersey Institute of TechnologyAlzheimer's Disease Neuroimaging Initiative
- Contributors
- Delwyn D Miller (Contributor) - University of Iowa, PsychiatryLaura L Boles-Ponto (Contributor) - University of Iowa, RadiologyHyungSub Shim (Contributor) - University of Iowa, NeurologyHristina K Koleva (Contributor) - University of Iowa, Psychiatry
- Resource Type
- Journal article
- Publication Details
- Scandinavial Journal of Statistics, Vol.50(4), pp.1667-1686
- Publisher
- Wiley
- DOI
- 10.1111/sjos.12660
- ISSN
- 0303-6898
- eISSN
- 1467-9469
- Grant note
- DOI: 10.13039/100000001, name: National Science Foundation, award: DMS 1736470, DMS 1764280, 1821157; DOI: 10.13039/100000893, name: Simons Foundation, award: 849413
- Language
- English
- Electronic publication date
- 05/24/2023
- Date published
- 12/2023
- Academic Unit
- Neurology; Psychiatry; Radiology
- Record Identifier
- 9984414754802771
Metrics
2 Record Views