# Neuron-Level Architecture Growth: A Controlled Evaluation for EEG Time-Series Decoding

arXiv: https://arxiv.org/abs/2609.33880 · HTML: https://mlsanity.com/paper/2609.33880/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 33 references that are in the map.
- First posted on arXiv 27 Sep 2026.

## Builds on (most cited first, 20 of 20)

- [EEGNet: A Compact Convolutional Network for EEG-based Brain-Computer Interfaces](https://mlsanity.com/paper/1611.08024/) (2016)
- [Net2Net: Accelerating Learning via Knowledge Transfer](https://mlsanity.com/paper/1511.05641/) (2015)
- [MOABB: Trustworthy algorithm benchmarking for BCIs](https://mlsanity.com/paper/1805.06427/) (2018)
- [GradMax: Growing Neural Networks using Gradient Information](https://mlsanity.com/paper/2201.05125/) (2022)
- [Splitting Steepest Descent for Growing Neural Architectures](https://mlsanity.com/paper/1910.02366/) (2019)
- [Firefly Neural Architecture Descent: a General Approach for Growing Neural Networks](https://mlsanity.com/paper/2102.08574/) (2021)
- [EEG Foundation Models: A Critical Review of Current Progress and Future Directions](https://mlsanity.com/paper/2507.11783/) (2025)
- [The largest EEG-based BCI reproducibility study for open science: the MOABB benchmark](https://mlsanity.com/paper/2404.15319/) (2024)
- [Are Large Brainwave Foundation Models Capable Yet? Insights from Fine-tuning](https://mlsanity.com/paper/2507.01196/) (2025)
- [A Systematic Evaluation of Euclidean Alignment with Deep Learning for EEG Decoding](https://mlsanity.com/paper/2401.10746/) (2024)
- [Growing Tiny Networks: Spotting Expressivity Bottlenecks and Fixing Them Optimally](https://mlsanity.com/paper/2405.19816/) (2024)
- [Evaluating the structure of cognitive tasks with transfer learning](https://mlsanity.com/paper/2308.02408/) (2023)
- [MI-BMInet: An Efficient Convolutional Neural Network for Motor Imagery Brain--Machine Interfaces with EEG Channel Selection](https://mlsanity.com/paper/2203.14592/) (2022)
- [Minimizing subject-dependent calibration for BCI with Riemannian transfer learning](https://mlsanity.com/paper/2111.12071/) (2021)
- [Adaptive Width Neural Networks](https://mlsanity.com/paper/2501.15889/) (2025)
- [Deep comparisons of Neural Networks from the EEGNet family](https://mlsanity.com/paper/2302.08797/) (2023)
- [When To Grow? A Fitting Risk-Aware Policy for Layer Growing in Deep Neural Networks](https://mlsanity.com/paper/2401.03104/) (2024)
- [Cross Task Neural Architecture Search for EEG Signal Classifications](https://mlsanity.com/paper/2210.06298/) (2022)
- [Growth strategies for arbitrary DAG neural architectures](https://mlsanity.com/paper/2501.12690/) (2025)
- [Towards a Network Expansion Approach for Reliable Brain-Computer Interface](https://mlsanity.com/paper/2411.11872/) (2024)

## Abstract (arXiv)

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.