English

Electromagnetic neural source imaging under sparsity constraints with SURE-based hyperparameter tuning

Image and Video Processing 2021-12-24 v1 Signal Processing Optimization and Control

Abstract

Estimators based on non-convex sparsity-promoting penalties were shown to yield state-of-the-art solutions to the magneto-/electroencephalography (M/EEG) brain source localization problem. In this paper we tackle the model selection problem of these estimators: we propose to use a proxy of the Stein's Unbiased Risk Estimator (SURE) to automatically select their regularization parameters. The effectiveness of the method is demonstrated on realistic simulations and 3030 subjects from the Cam-CAN dataset. To our knowledge, this is the first time that sparsity promoting estimators are automatically calibrated at such a scale. Results show that the proposed SURE approach outperforms cross-validation strategies and state-of-the-art Bayesian statistics methods both computationally and statistically.

Keywords

Cite

@article{arxiv.2112.12178,
  title  = {Electromagnetic neural source imaging under sparsity constraints with SURE-based hyperparameter tuning},
  author = {Pierre-Antoine Bannier and Quentin Bertrand and Joseph Salmon and Alexandre Gramfort},
  journal= {arXiv preprint arXiv:2112.12178},
  year   = {2021}
}
R2 v1 2026-06-24T08:28:36.884Z