# On-device Learning of EEGNet-based Network For Wearable Motor Imagery Brain-Computer Interface

arXiv: https://arxiv.org/abs/2409.00083 · HTML: https://mlsanity.com/paper/2409.00083/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 48 references that are in the map.
- First posted on arXiv 25 Aug 2024.

## Builds on (most cited first, 16 of 16)

- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [EEGNet: A Compact Convolutional Network for EEG-based Brain-Computer Interfaces](https://mlsanity.com/paper/1611.08024/) (2016)
- [Towards Stable Test-Time Adaptation in Dynamic Wild World](https://mlsanity.com/paper/2302.12400/) (2023)
- [TinyOL: TinyML with Online-Learning on Microcontrollers](https://mlsanity.com/paper/2103.08295/) (2021)
- [SoTTA: Robust Test-Time Adaptation on Noisy Data Streams](https://mlsanity.com/paper/2310.10074/) (2023)
- [A TinyML Platform for On-Device Continual Learning with Quantized Latent Replays](https://mlsanity.com/paper/2110.10486/) (2021)
- [Retina : Low-Power Eye Tracking with Event Camera and Spiking Hardware](https://mlsanity.com/paper/2312.00425/) (2023)
- [Optimization-Free Test-Time Adaptation for Cross-Person Activity Recognition](https://mlsanity.com/paper/2310.18562/) (2023)
- [An Accurate EEGNet-based Motor-Imagery Brain-Computer Interface for Low-Power Edge Computing](https://mlsanity.com/paper/2004.00077/) (2020)
- [TinyTracker: Ultra-Fast and Ultra-Low-Power Edge Vision In-Sensor for Gaze Estimation](https://mlsanity.com/paper/2307.07813/) (2023)
- [Ultra-Efficient On-Device Object Detection on AI-Integrated Smart Glasses with TinyissimoYOLO](https://mlsanity.com/paper/2311.01057/) (2023)
- [On-Device Domain Learning for Keyword Spotting on Low-Power Extreme Edge Embedded Systems](https://mlsanity.com/paper/2403.10549/) (2024)
- [M3BAT: Unsupervised Domain Adaptation for Multimodal Mobile Sensing with Multi-Branch Adversarial Training](https://mlsanity.com/paper/2404.17391/) (2024)
- [Lifelong Adaptive Machine Learning for Sensor-based Human Activity Recognition Using Prototypical Networks](https://mlsanity.com/paper/2203.05692/) (2022)
- [ColibriUAV: An Ultra-Fast, Energy-Efficient Neuromorphic Edge Processing UAV-Platform with Event-Based and Frame-Based Cameras](https://mlsanity.com/paper/2305.18371/) (2023)
- [Incremental Online Learning Algorithms Comparison for Gesture and Visual Smart Sensors](https://mlsanity.com/paper/2209.00591/) (2022)

## Built on by (most cited first, 3 of 3)

- [TinierHAR: Towards Ultra-Lightweight Deep Learning Models for Efficient Human Activity Recognition on Edge Devices](https://mlsanity.com/paper/2507.07949/) (2025)
- [BrainStack: Neuro-MoE with Functionally Guided Expert Routing for EEG-Based Language Decoding](https://mlsanity.com/paper/2601.21148/) (2026)
- [NEXUS-MI: Communication-Aware Federated Personalization for Gateway-Coordinated Motor-Imagery Brain-Computer Interfaces](https://mlsanity.com/paper/2609.09786/) (2026)

## Abstract (arXiv)

Electroencephalogram (EEG)-based Brain-Computer Interfaces (BCIs) have garnered significant interest across various domains, including rehabilitation and robotics. Despite advancements in neural network-based EEG decoding, maintaining performance across diverse user populations remains challenging due to feature distribution drift. This paper presents an effective approach to address this challenge by implementing a lightweight and efficient on-device learning engine for wearable motor imagery recognition. The proposed approach, applied to the well-established EEGNet architecture, enables real-time and accurate adaptation to EEG signals from unregistered users. Leveraging the newly released low-power parallel RISC-V-based processor, GAP9 from Greeenwaves, and the Physionet EEG Motor Imagery dataset, we demonstrate a remarkable accuracy gain of up to 7.31% with respect to the baseline with a memory footprint of 15.6 KByte. Furthermore, by optimizing the input stream, we achieve enhanced real-time performance without compromising inference accuracy. Our tailored approach exhibits inference time of 14.9 ms and 0.76 mJ per single inference and 20 us and 0.83 uJ per single update during online training. These findings highlight the feasibility of our method for edge EEG devices as well as other battery-powered wearable AI systems suffering from subject-dependant feature distribution drift.