Towards Privacy Preserving BCIs: Profiling the Feasibility of Federated Learning for Motor Imagery Brain-Computer Interfaces
Bibliographic record
Abstract
Brain-computer interfaces (BCIs) are positioned to help individuals with physical disabilities, yet training data for these systems are time-consuming and expensive to collect. BCIs could thus benefit from cross-user data sharing, but as neurophysiological signals used in BCIs are personal health information, the protection of user privacy and data sovereignty will be of the utmost importance for end-user BCI applications. Federated learning is a novel privacy-preserving machine learning technique that decentralizes training, leaving end-user data on the device/institution of origin. In this work, we explored the feasibility of federated learning for BCIs, profiling convergence and performance for two federated learning algorithms (FedAvg and FedDC) on both identically and independently distributed (IID) and non-IID partitions of data. Federated learning for a 4-class motor imagery BCI decoding task was found to be feasible, although came at a cost of reduced performance (longer convergence rate and reduced accuracy). The FedDC algorithm, which introduces drift correction for heterogeneous data, well out-performed the FedAvg algorithm. Larger datasets with more subjects and data per subject would be beneficial for further investigations, and future work should explore federated transfer learning techniques for combating inter-subject data heterogeneity and improving global model performance.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.010 | 0.051 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".