A Unified Framework for Domain Adaptation in EEG-based Brain-Computer Interfaces.

R implementation of DA4BCI — a comprehensive toolkit of domain adaptation methods, distance metrics, and evaluation tools for EEG-based BCI applications. The package provides a unified interface to align EEG distributions across sessions or subjects, mitigating distributional shift and improving model robustness.

Looking for the Python version? See DA4BCI-Python.

Requires R >= 3.5.0.

install.packages("remotes")
remotes::install_github("Yiming-S/DA4BCI", force = TRUE)

library(DA4BCI)

# Simulate EEG-like source and target features
set.seed(1)
source_data <- matrix(rnorm(100 * 20), 100, 20)
target_data <- matrix(rnorm(100 * 20) + 0.5, 100, 20)

# Apply domain adaptation (unified interface)
res <- domain_adaptation(
  source_data,
  target_data,
  method  = "coral",
  control = list(lambda = 1e-5)
)

adapted_source <- res$weighted_source_data
adapted_target <- res$target_data

# Quantify distribution alignment
distanceSummary(adapted_source, adapted_target, format = "table")

All methods are dispatched through the unified domain_adaptation() function or can be called directly.

DA4BCI includes a set of distance metrics for quantitatively assessing the effect of adaptation:

• Euclidean Distance Matrix — efficient pairwise distances between datasets.
• Wasserstein Distance — minimal transport cost between distributions.
• Maximum Mean Discrepancy (MMD) — kernel-based discrepancy, sensitive to subtle shifts.
• Energy Distance — empirical-distribution discrepancy from pairwise distances.
• Mahalanobis Distance — whitening-aware distance with optional shrinkage covariance.

The distanceSummary() function bundles these into a single table for quick before/after comparison.

distanceSummary(
  source_data,
  target_data,
  include = c("MMD", "Energy", "Wasserstein", "Mahalanobis"),
  format  = "table"
)

For a practical method-selection guide organized by shift type, data representation, supervision level, and risk, see ALGORITHM_SELECTION_GUIDE.md.

Quick rules of thumb:

• Start with SA and CORAL as fast linear baselines.
• If covariance structure dominates, move to PT and ART; keep RD as a lightweight baseline.
• If mismatch looks nonlinear or cluster-wise, try OT.
• If you have reliable source labels and need class-aware refinement, try M3D.

devtools::test()

MIT

• Yiming Shen — yiming.shen001@umb.edu
• David Degras — david.degras@umb.edu