Functional Neural Networks: Shift invariant models for functional data with applications to EEG classification
This work addresses the need for shift-invariant models in functional data analysis, with applications to EEG classification, representing an incremental advancement by extending existing neural network methods to functional data.
The authors tackled the problem of detecting signals independently of their position in smooth functional data by introducing functional neural networks (FNNs), which outperformed a benchmark model in accuracy and successfully classified EEG data.
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.