# Structure-Preserving Correction Learning for Sparse Bayesian Inference in Brain Source Imaging

arXiv: https://arxiv.org/abs/2606.07196 · HTML: https://mlsanity.com/paper/2606.07196/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 32 references that are in the map.
- First posted on arXiv 5 Jun 2026.

## Builds on (most cited first, 8 of 8)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Algorithm Unrolling: Interpretable, Efficient Deep Learning for Signal and Image Processing](https://mlsanity.com/paper/1912.10557/) (2019)
- [Deeply-Supervised Nets](https://mlsanity.com/paper/1409.5185/) (2014)
- [Deep Unfolding: Model-Based Inspiration of Novel Deep Architectures](https://mlsanity.com/paper/1409.2574/) (2014)
- [MEG Source Localization via Deep Learning](https://mlsanity.com/paper/2012.00588/) (2020)
- [Efficient Hierarchical Bayesian Inference for Spatio-temporal Regression Models in Neuroimaging](https://mlsanity.com/paper/2111.01692/) (2021)
- [Sparse Bayesian Learning for DOA Estimation in Heteroscedastic Noise](https://mlsanity.com/paper/1711.03847/) (2017)
- [Enhancing Brain Source Reconstruction by Initializing 3D Neural Networks with Physical Inverse Solutions](https://mlsanity.com/paper/2411.00143/) (2024)

## Built on by (most cited first, 1 of 1)

- [Decoupling Time and Space: A Temporally Conditioned Refinement for EEG Source Imaging](https://mlsanity.com/paper/2610.06726/) (2026)

## Abstract (arXiv)

Classical sparse Type-II Bayesian methods for M/EEG brain imaging support joint estimation of source and noise hyperparameters, but rely on fixed iterative update rules. Although these updates are principled and interpretable, their dynamics are not data-driven. We propose to learn the update mechanism itself while preserving the underlying Bayesian structure by unfolding a classical joint hyperparameter-learning solver into a trainable neural network architecture whose layers mirror the original iterations. The resulting framework is initialized to recover the classical solver exactly before training and is enriched through progressively more expressive residual correction learning mechanisms, ranging from learnable biases to adaptive multilayer perceptron and attention-based contextual refinements. In this way, training does not replace Bayesian inference with a black-box predictor, but instead learns structured correction terms while retaining the interpretability and model-based character of the original update dynamics. Structured correction learning therefore aims to improve empirical reconstruction performance without replacing the original model-based inference mechanism. Experimental results show that the learned correction variants improve reconstruction performance and convergence behavior over the baseline unfolded solver while preserving its algorithmic transparency.