Pretraining for Sample-Efficient Neural Interfaces
Abstract
Brain-computer interfaces (BCIs) decode neural activity to restore lost function. Typically, training a high-performance neural decoder requires a large labeled dataset to be collected from every new subject. One way to reduce the labeled data cost is self-supervised pretraining, which learns general neural representations from unlabeled recordings that accumulate across subjects. However, for intracranial electroencephalography (iEEG) recordings, self-supervised learning has been challenging due to differences in contact placement and neuroanatomy between subjects. We propose MAPA, an otherwise vanilla masked autoencoder with two spatial encodings, an anatomical region embedding and a relative positional encoding, which together enable it to learn neural representations that transfer to unseen subjects and across various tasks. MAPA sets a new state of the art across all three regimes of the Neuroprobe benchmark without fine-tuning: within-session, cross-session, and cross-subject. In the cross-subject regime, a linear probe on MAPA's features needs only ∼164 labeled trials to reach the accuracy that takes 3,500 without pretraining. Our results show that self-supervised pretraining can scale across heterogeneous iEEG recordings and reduce the labeled data needed for accurate decoding in new subjects.
Related
- BrainWideBench: Benchmarking large-scale pretraining and across-animal transfer in multi-region neural recordings
- Generalizable neural decoding for clinically scalable brain-computer interfaces.
- Slow-Fast Brain-Computer Interfaces: Preventing Neuroadaptive Overfitting in AI-Mediated Neural Interfaces
- Assistive algorithms influence neural representations in motor brain-computer interfaces
- Functional Hydrogels for Wearable and Implantable Neural Interfaces: Materials, Properties, and Applications.
BibTeX
@misc{tang2026pretraining,
title = {{Pretraining for Sample-Efficient Neural Interfaces}},
author = {Ben Tang and Zachary Spalding and Gregory B. Cogan},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.13507},
eprint = {2609.13507},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.13507},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Pretraining for Sample-Efficient Neural Interfaces
AU - Ben Tang
AU - Zachary Spalding
AU - Gregory B. Cogan
PY - 2026
DA - 2026/09/11
PB - arXiv
DO - 10.48550/arxiv.2609.13507
UR - https://arxiv.org/abs/2609.13507
AB - Brain-computer interfaces (BCIs) decode neural activity to restore lost function. Typically, training a high-performance neural decoder requires a large labeled dataset to be collected from every new subject. One way to reduce the labeled data cost is self-supervised pretraining, which learns general neural representations from unlabeled recordings that accumulate across subjects. However, for intracranial electroencephalography (iEEG) recordings, self-supervised learning has been challenging due to differences in contact placement and neuroanatomy between subjects. We propose MAPA, an otherwise vanilla masked autoencoder with two spatial encodings, an anatomical region embedding and a relative positional encoding, which together enable it to learn neural representations that transfer to unseen subjects and across various tasks. MAPA sets a new state of the art across all three regimes of the Neuroprobe benchmark without fine-tuning: within-session, cross-session, and cross-subject. In the cross-subject regime, a linear probe on MAPA's features needs only ∼164 labeled trials to reach the accuracy that takes 3,500 without pretraining. Our results show that self-supervised pretraining can scale across heterogeneous iEEG recordings and reduce the labeled data needed for accurate decoding in new subjects.
ER -