The Parameterized Complexity of Finding Concise Local Explanations

Sebastian Ordyniak, Giacomo Paesani, Stefan Szeider

IJCAI 2023 pp. 3312-3320

doi:10.24963/IJCAI.2023/369 /ijcai/2023/ordyniak2023ijcai-parameterized/

Abstract

We consider the computational problem of finding a smallest local explanation (anchor) for classifying a given feature vector (example) by a black-box model. After showing that the problem is NP-hard in general, we study various natural restrictions of the problem in terms of problem parameters to see whether these restrictions make the problem fixed-parameter tractable or not. We draw a detailed and systematic complexity landscape for combinations of parameters, including the size of the anchor, the size of the anchor's coverage, and parameters that capture structural aspects of the problem instance, including rank-width, twin-width, and maximum difference.

PDF IJCAI Semantic Scholar

Cite

Text

Ordyniak et al. "The Parameterized Complexity of Finding Concise Local Explanations." International Joint Conference on Artificial Intelligence, 2023. doi:10.24963/IJCAI.2023/369

Markdown

[Ordyniak et al. "The Parameterized Complexity of Finding Concise Local Explanations." International Joint Conference on Artificial Intelligence, 2023.](https://mlanthology.org/ijcai/2023/ordyniak2023ijcai-parameterized/) doi:10.24963/IJCAI.2023/369

BibTeX

@inproceedings{ordyniak2023ijcai-parameterized,
  title     = {{The Parameterized Complexity of Finding Concise Local Explanations}},
  author    = {Ordyniak, Sebastian and Paesani, Giacomo and Szeider, Stefan},
  booktitle = {International Joint Conference on Artificial Intelligence},
  year      = {2023},
  pages     = {3312-3320},
  doi       = {10.24963/IJCAI.2023/369},
  url       = {https://mlanthology.org/ijcai/2023/ordyniak2023ijcai-parameterized/}
}