Functional Neural Networks: Shift invariant models for functional data with applications to EEG classification

Florian Heinrichs, Mavin Heim, Corinna Weber
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:12866-12881, 2023.

Abstract

It is desirable for statistical models to detect signals of interest independently of their position. If the data is generated by some smooth process, this additional structure should be taken into account. We introduce a new class of neural networks that are shift invariant and preserve smoothness of the data: functional neural networks (FNNs). For this, we use methods from functional data analysis (FDA) to extend multi-layer perceptrons and convolutional neural networks to functional data. We propose different model architectures, show that the models outperform a benchmark model from FDA in terms of accuracy and successfully use FNNs to classify electroencephalography (EEG) data.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-heinrichs23a, title = {Functional Neural Networks: Shift invariant models for functional data with applications to {EEG} classification}, author = {Heinrichs, Florian and Heim, Mavin and Weber, Corinna}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {12866--12881}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/heinrichs23a/heinrichs23a.pdf}, url = {https://proceedings.mlr.press/v202/heinrichs23a.html}, abstract = {It is desirable for statistical models to detect signals of interest independently of their position. If the data is generated by some smooth process, this additional structure should be taken into account. We introduce a new class of neural networks that are shift invariant and preserve smoothness of the data: functional neural networks (FNNs). For this, we use methods from functional data analysis (FDA) to extend multi-layer perceptrons and convolutional neural networks to functional data. We propose different model architectures, show that the models outperform a benchmark model from FDA in terms of accuracy and successfully use FNNs to classify electroencephalography (EEG) data.} }
Endnote
%0 Conference Paper %T Functional Neural Networks: Shift invariant models for functional data with applications to EEG classification %A Florian Heinrichs %A Mavin Heim %A Corinna Weber %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-heinrichs23a %I PMLR %P 12866--12881 %U https://proceedings.mlr.press/v202/heinrichs23a.html %V 202 %X It is desirable for statistical models to detect signals of interest independently of their position. If the data is generated by some smooth process, this additional structure should be taken into account. We introduce a new class of neural networks that are shift invariant and preserve smoothness of the data: functional neural networks (FNNs). For this, we use methods from functional data analysis (FDA) to extend multi-layer perceptrons and convolutional neural networks to functional data. We propose different model architectures, show that the models outperform a benchmark model from FDA in terms of accuracy and successfully use FNNs to classify electroencephalography (EEG) data.
APA
Heinrichs, F., Heim, M. & Weber, C.. (2023). Functional Neural Networks: Shift invariant models for functional data with applications to EEG classification. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:12866-12881 Available from https://proceedings.mlr.press/v202/heinrichs23a.html.

Related Material