English

Compositionality Unlocks Deep Interpretable Models

Machine Learning 2025-04-04 v1

Abstract

We propose χ\chi-net, an intrinsically interpretable architecture combining the compositional multilinear structure of tensor networks with the expressivity and efficiency of deep neural networks. χ\chi-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.

Keywords

Cite

@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}
}
R2 v1 2026-06-28T22:45:26.764Z