We propose χ-net, an intrinsically interpretable architecture combining the compositional multilinear structure of tensor networks with the expressivity and efficiency of deep neural networks. χ-nets retain equal accuracy compared to their baseline counterparts. Our novel, efficient diagonalisation algorithm, ODT, reveals linear low-rank structure in a multilayer SVHN model. We leverage this toward formal weight-based interpretability and model compression.
@article{arxiv.2504.02667,
title = {Compositionality Unlocks Deep Interpretable Models},
author = {Thomas Dooms and Ward Gauderis and Geraint A. Wiggins and Jose Oramas},
journal= {arXiv preprint arXiv:2504.02667},
year = {2025}
}