MTEEG: A Multi-Task Learning Framework for Enhanced Electroencephalography Analysis Using Low-Rank Adaptation
Keywords: EEG, brain-computer interface, multi-task learning
Abstract: Electroencephalography (EEG) analysis using deep learning has traditionally placed a strong emphasis on models that are custom-built and optimized for specific datasets. Several recent research utilize self-supervised learning to extract generic representations from massive amounts of unlabeled EEG data. The pre-trained models are then fine-tuned on each downstream dataset independently, demonstrating promising results. However, in practical applications involving multiple tasks, utilizing a separate model for each is not ideal regarding computational and spatial cost. In this study, we go one step further and explore the simultaneous adaptation of a pre-trained model to multiple different tasks. The EEG signals exhibit significant heterogeneity due to their collection from various subjects using diverse devices and experimental setups, resulting in potential conflicts among different tasks that impede joint optimization. To tackle this challenge, we propose MTEEG, a multi-task EEG recognition framework which incorporates a task-agnostic temporal encoder and task-specific low-rank adaptation modules to disentangle the parameter space, facilitating both task interaction and specification. Experiments show that MTEEG surpasses other multi-task methods and performs on par with state-of-the-art single-task methods on abnormal detection, event type classification, emotion recognition, seizure detection, sleep stage classification and motor imagery classification after being tuned jointly on six publicly available datasets. MTEEG shows the potential of multi-task EEG recognition and promotes the development of general-purpose brain-computer interfaces in the future. The source code will be released.
Primary Area: applications to neuroscience & cognitive science
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 12753
Loading