ML Anthology

Interpretable Regional Descriptors: Hyperbox-Based Local Explanations

Susanne Dandl, Giuseppe Casalicchio, Bernd Bischl, Ludwig Bothmann
ECML-PKDD 2023 pp. 479-495
doi:10.1007/978-3-031-43418-1_29 /ecmlpkdd/2023/dandl2023ecmlpkdd-interpretable/

Abstract

This work introduces interpretable regional descriptors, or IRDs, for local, model-agnostic interpretations. IRDs are hyperboxes that describe how an observation’s feature values can be changed without affecting its prediction. They justify a prediction by providing a set of “even if” arguments (semi-factual explanations), and they indicate which features affect a prediction and whether pointwise biases or implausibilities exist. A concrete use case shows that this is valuable for both machine learning modelers and persons subject to a decision. We formalize the search for IRDs as an optimization problem and introduce a unifying framework for computing IRDs that covers desiderata, initialization techniques, and a post-processing method. We show how existing hyperbox methods can be adapted to fit into this unified framework. A benchmark study compares the methods based on several quality measures and identifies two strategies to improve IRDs.

PDF ECML-PKDD Semantic Scholar

Cite

Text

Dandl et al. "Interpretable Regional Descriptors: Hyperbox-Based Local Explanations." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2023. doi:10.1007/978-3-031-43418-1_29

Markdown

[Dandl et al. "Interpretable Regional Descriptors: Hyperbox-Based Local Explanations." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2023.](https://mlanthology.org/ecmlpkdd/2023/dandl2023ecmlpkdd-interpretable/) doi:10.1007/978-3-031-43418-1_29

BibTeX

@inproceedings{dandl2023ecmlpkdd-interpretable,
  title     = {{Interpretable Regional Descriptors: Hyperbox-Based Local Explanations}},
  author    = {Dandl, Susanne and Casalicchio, Giuseppe and Bischl, Bernd and Bothmann, Ludwig},
  booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
  year      = {2023},
  pages     = {479-495},
  doi       = {10.1007/978-3-031-43418-1_29},
  url       = {https://mlanthology.org/ecmlpkdd/2023/dandl2023ecmlpkdd-interpretable/}
}