# Multilayer Perceptron with Functional Inputs: an Inverse Regression   Approach

**Authors:** Louis Ferr\'e (GRIMM), Nathalie Villa (GRIMM)

arXiv: 0705.0211 · 2007-05-23

## TL;DR

This paper introduces a novel approach combining functional sliced inverse regression with neural networks for regression and classification tasks involving functional data, providing theoretical consistency and real-world validation.

## Contribution

It develops a regularized version of functional sliced inverse regression integrated with neural networks, advancing functional data analysis methods.

## Key findings

- The method achieves competitive accuracy on real-life datasets.
- Consistency results support the theoretical validity of the approach.
- The approach effectively reduces dimensionality in functional data analysis.

## Abstract

Functional data analysis is a growing research field as more and more practical applications involve functional data. In this paper, we focus on the problem of regression and classification with functional predictors: the model suggested combines an efficient dimension reduction procedure [functional sliced inverse regression, first introduced by Ferr\'e & Yao (Statistics, 37, 2003, 475)], for which we give a regularized version, with the accuracy of a neural network. Some consistency results are given and the method is successfully confronted to real-life data.

## Full text

Tomesphere does not host the full text of arXiv papers. Read it on arXiv: https://arxiv.org/abs/0705.0211

## References

31 references — full list in the complete paper: https://tomesphere.com/paper/0705.0211/full.md

---
Source: https://tomesphere.com/paper/0705.0211