ICML 2022spotlight17 citations

Interpretable Neural Networks with Frank-Wolfe: Sparse Relevance Maps and Relevance Orderings

Jan Macdonald, Mathieu E. Besançon, Sebastian Pokutta

Abstract

We study the effects of constrained optimization formulations and Frank-Wolfe algorithms for obtaining interpretable neural network predictions. Reformulating the Rate-Distortion Explanations (RDE) method for relevance attribution as a constrained optimization problem provides precise control over the sparsity of relevance maps. This enables a novel multi-rate as well as a relevance-ordering variant of RDE that both empirically outperform standard RDE and other baseline methods in a well-established comparison test. We showcase several deterministic and stochastic variants of the Frank-Wolfe algorithm and their effectiveness for RDE.

BibTeX
@InProceedings{pmlr-v162-macdonald22a,
  title = 	 {Interpretable Neural Networks with Frank-Wolfe: Sparse Relevance Maps and Relevance Orderings},
  author =       {Macdonald, Jan and Besan{\c{c}}on, Mathieu E. and Pokutta, Sebastian},
  booktitle = 	 {Proceedings of the 39th International Conference on Machine Learning},
  pages = 	 {14699--14716},
  year = 	 {2022},
  editor = 	 {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan},
  volume = 	 {162},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {17--23 Jul},
  publisher =    {PMLR},
  pdf = 	 {https://proceedings.mlr.press/v162/macdonald22a/macdonald22a.pdf},
  url = 	 {https://proceedings.mlr.press/v162/macdonald22a.html},
  abstract = 	 {We study the effects of constrained optimization formulations and Frank-Wolfe algorithms for obtaining interpretable neural network predictions. Reformulating the Rate-Distortion Explanations (RDE) method for relevance attribution as a constrained optimization problem provides precise control over the sparsity of relevance maps. This enables a novel multi-rate as well as a relevance-ordering variant of RDE that both empirically outperform standard RDE and other baseline methods in a well-established comparison test. We showcase several deterministic and stochastic variants of the Frank-Wolfe algorithm and their effectiveness for RDE.}
}