Neuron level growth improves some EEG convolutional decoders accuracy
Neuron-Level Architecture Growth: A Controlled Evaluation for EEG Time-Series Decoding
Human-Computer InteractionMachine LearningNeural and Evolutionary Computing
Summary
Many EEG decoding models are built with a fixed number of neurons, which might not be the best size for every dataset. The authors tested methods that add neurons during training only where it helps the model improve the most. They found that for some simpler brain activity decoders, growing the network helped improve accuracy and reduced the number of parameters needed. However, for deeper models, growing neurons did not improve performance as expected. This suggests that carefully deciding where to add neurons matters for improving EEG decoding.
What this means in practice
- •For brain computer interface developers: Improve EEG motor imagery decoders by dynamically growing neurons where it most reduces error for more accurate and compact models.
- •For neural signal processing engineers: Design EEG decoders that start small and grow neuron layers guided by a data-driven threshold to optimize decoding on new subjects.
Authors
Adam Mounir, Stella Douka, Arnault H. Caillet, Bruno Aristimunha, Sylvain Chevallier
Abstract
Convolutional EEG decoders are trained at a fixed width, usually set by their authors on other data. Growing methods add neurons during training where the loss could decrease the most, but whether they improve compared to a reference width is untested on EEG. Here, we grow three convolutional backbones on 12 motor-imagery datasets under three protocols and compare each with its reference model per subject. The growing ShallowFBCSPNet scores 2.9 points above its reference model with only half the parameters (0.57x), SCCNet changes by at most 1.2 points. Deep4Net growing models show decreased accuracy, but they require adaptation that prevent to compare faithfully the results. These differences follow the selection step, which keeps a candidate neuron relying on a dynamic threshold from singular values decomposition. Overall, these results suggest that growth helps when its criterion can rank the candidate neurons, and that the rate of skipped neuron addition tells where a decoder can be grown small from scratch.