Non-linear Functional Modeling using Neural Networks

Abstract

We introduce a new class of non-linear models for functional data based on neural networks. Deep learning has been very successful in non-linear modeling, but there has been little work done in the functional data setting. We propose two variations of our framework: a functional neural network with continuous hidden layers, called the Functional Direct Neural Network (FDNN), and a second version that utilizes basis expansions and continuous hidden layers, called the Functional Basis Neural Network (FBNN). Both are designed explicitly to exploit the structure inherent in functional data. To fit these models we derive a functional gradient based optimization algorithm. The effectiveness of the proposed methods in handling complex functional models is demonstrated by comprehensive simulation studies and real data examples.

arXiv comment3 figures, 10 tables (including supplementary material), 14 pages (including supplementary material)

References in corpus (5)

• DRAW: A Recurrent Neural Network For Image Generation
• A Dual-Stage Attention-Based Recurrent Neural Network for Time Series Prediction
• Functional additive regression
• Remaining Useful Life Estimation Using Functional Data Analysis
• Theoretical Properties of Projection Based Multilayer Perceptrons with Functional Inputs

Cited by in corpus (3)

• An ensemble of convolution-based methods for fault detection using vibration signals
• Modern Non-Linear Function-on-Function Regression
• Spatially Varying Deep Functional Neural Network: Application in Large-Scale Crop Yield Prediction