# End-to-end learnable EEG channel selection for deep neural networks with Gumbel-softmax

arXiv: https://arxiv.org/abs/2102.09050 · HTML: https://mlsanity.com/paper/2102.09050/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 39 references that are in the map.
- First posted on arXiv 11 Feb 2021.

## Builds on (most cited first, 12 of 12)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Categorical Reparameterization with Gumbel-Softmax](https://mlsanity.com/paper/1611.01144/) (2016)
- [The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables](https://mlsanity.com/paper/1611.00712/) (2016)
- [EEGNet: A Compact Convolutional Network for EEG-based Brain-Computer Interfaces](https://mlsanity.com/paper/1611.08024/) (2016)
- [Learning Sparse Neural Networks through L0 Regularization](https://mlsanity.com/paper/1712.01312/) (2017)
- [Deep learning-based electroencephalography analysis: a systematic review](https://mlsanity.com/paper/1901.05498/) (2019)
- [Concrete Autoencoders for Differentiable Feature Selection and Reconstruction](https://mlsanity.com/paper/1901.09346/) (2019)
- [Group Sparse Regularization for Deep Neural Networks](https://mlsanity.com/paper/1607.00485/) (2016)
- [Gradient Estimation with Stochastic Softmax Tricks](https://mlsanity.com/paper/2006.08063/) (2020)
- [FsNet: Feature Selection Network on High-dimensional Biological Data](https://mlsanity.com/paper/2001.08322/) (2020)
- [An LSTM Based Architecture to Relate Speech Stimulus to EEG](https://mlsanity.com/paper/2002.10988/) (2020)

## Built on by (most cited first, 12 of 12)

- [Symbolic--KAN: Kolmogorov-Arnold Networks with Discrete Symbolic Structure for Interpretable Learning](https://mlsanity.com/paper/2603.23854/) (2026)
- [Indirectly Parameterized Concrete Autoencoders](https://mlsanity.com/paper/2403.00563/) (2024)
- [Enhancing Low-Density EEG-Based Brain-Computer Interfaces with Similarity-Keeping Knowledge Distillation](https://mlsanity.com/paper/2212.03329/) (2022)
- [Unsupervised Feature Selection to Identify Important ICD-10 Codes for Machine Learning: A Case Study on a Coronary Artery Disease Patient Cohort](https://mlsanity.com/paper/2303.14303/) (2023)
- [Knowledge-Distilled Graph Neural Networks for Personalized Epileptic Seizure Detection](https://mlsanity.com/paper/2304.06038/) (2023)
- [A distributed neural network architecture for dynamic sensor selection with application to bandwidth-constrained body-sensor networks](https://mlsanity.com/paper/2308.08379/) (2023)
- [Conditional Gumbel-Softmax for constrained feature selection with application to node selection in wireless sensor networks](https://mlsanity.com/paper/2406.01162/) (2024)
- [Bandwidth-efficient distributed neural network architectures with application to body sensor networks](https://mlsanity.com/paper/2210.07750/) (2022)
- [LGL-BCI: A Motor-Imagery-Based Brain-Computer Interface with Geometric Learning](https://mlsanity.com/paper/2310.08051/) (2023)
- [EEGChaT: A Transformer-Based Modular Channel Selector for SEEG Analysis](https://mlsanity.com/paper/2510.13592/) (2025)
- [FlexAct: Why Learn when you can Pick?](https://mlsanity.com/paper/2601.06441/) (2026)
- [Position: AI for Science Should Treat Measurement-to-Dataset Pipelines as Inference Components](https://mlsanity.com/paper/2605.24558/) (2026)

## Abstract (arXiv)

Many electroencephalography (EEG) applications rely on channel selection methods to remove the least informative channels, e.g., to reduce the amount of electrodes to be mounted, to decrease the computational load, or to reduce overfitting effects and improve performance. Wrapper-based channel selection methods aim to match the channel selection step to the target model, yet they require to re-train the model multiple times on different candidate channel subsets, which often leads to an unacceptably high computational cost, especially when said model is a (deep) neural network. To alleviate this, we propose a framework to embed the EEG channel selection in the neural network itself to jointly learn the network weights and optimal channels in an end-to-end manner by traditional backpropagation algorithms. We deal with the discrete nature of this new optimization problem by employing continuous relaxations of the discrete channel selection parameters based on the Gumbel-softmax trick. We also propose a regularization method that discourages selecting channels more than once. This generic approach is evaluated on two different EEG tasks: motor imagery brain-computer interfaces and auditory attention decoding. The results demonstrate that our framework is generally applicable, while being competitive with state-of-the art EEG channel selection methods, tailored to these tasks.