School of Mathematical and Statistical Sciences Faculty Publications and Presentations
Document Type
Article
Publication Date
2-19-2024
Abstract
In this paper, we show how convolutional neural networks (CNNs) can be used in regression and classification learning problems for noisy and non-noisy functional data (FD). The main idea is to transform the FD into a 28 by 28 image. We use a specific but typical architecture of a CNN to perform all the regression exercises of parameter estimation and functional form classification. First, we use some functional case studies of FD with and without random noise to showcase the strength of the new method. In particular, we use it to estimate exponential growth and decay rates, the bandwidths of sine and cosine functions, and the magnitudes and widths of curve peaks. We also use it to classify the monotonicity and curvatures of FD, the algebraic versus exponential growth, and the number of peaks of FD. Second, we apply the same CNNs to Lyapunov exponent estimation in noisy and non-noisy chaotic data, in estimating rates of disease transmission from epidemic curves, and in detecting the similarity of drug dissolution profiles. Finally, we apply the method to real-life data to detect Parkinson's disease patients in a classification problem. We performed ablation analysis and compared the new method with other commonly used neural networks for FD and showed that it outperforms them in all applications. Although simple, the method shows high accuracy and is promising for future use in engineering and medical applications.
Recommended Citation
Galarza, J., and T. Oraby. 2024. “Functional Data Learning Using Convolutional Neural Networks.” Machine Learning: Science and Technology 5 (1): 015030. https://doi.org/10.1088/2632-2153/ad2627
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Publication Title
Machine Learning: Science and Technology
DOI
10.1088/2632-2153/ad2627
Comments
Copyright the Authors.
Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 license. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.