Supervised Deep Multimodal Matrix Factorization for Interpretable Brain Network Analysis

TL;DR

SD3MF is a supervised deep matrix factorization framework for multimodal brain network analysis that produces interpretable features and outperforms existing deep learning models.

Contribution

It extends SNMTF to a supervised, multimodal setting with deep hierarchical factorizations and interpretable community-level features.

Findings

SD3MF outperforms CNNs and GNNs on brain datasets.

The model provides biologically interpretable features.

Adaptive weights enable effective multimodal data fusion.

Abstract

We present Supervised Deep Multimodal Matrix Factorization (SD3MF), an interpretable framework for integrative brain network analysis that generalizes Symmetric Nonnegative Matrix Tri-Factorization (SNMTF) from unsupervised single-graph clustering to supervised prediction over populations of multimodal graphs. SD3MF learns deep hierarchical factorizations for each modality together with a shared latent representation that aligns subjects across views. An encoder-decoder formulation jointly optimizes graph reconstruction and supervised prediction, while adaptive weights enable data-driven multimodal fusion. By representing each subject through community-level interaction matrices, the model yields interpretable and discriminative features. Experiments on multimodal connectome datasets show that SD3MF consistently outperforms strong deep learning baselines such as CNNs and GNNs, while enabling biologically interpretable insights. Code for reproducibility is available at: https://github.com/amjadseyedi/SD3MF.