Metraipolis › archive › dc_fb066441108a
Evaluating Explanation Methods by the Predictors They Induce — arXiv:2609.20058
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.20058
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-10T04:08:51.120Z
- SHA-256 of the archived text
95ed3c7469a1af4e5ed15fd4da66f9279a9397fbf0c5bebde37380acc66a167f- Stable URL of this page
https://metraipolis.com/documento/dc_fb066441108a
How to cite
Evaluating Explanation Methods by the Predictors They Induce — arXiv:2609.20058 — arXiv (en), https://arxiv.org/abs/2609.20058 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fb066441108a (Metraipolis archive dc_fb066441108a, sha256 95ed3c7469a1af4e5ed15fd4da66f9279a9397fbf0c5bebde37380acc66a167f)
BibTeX
@misc{dcfb066441108a,
title = {Evaluating Explanation Methods by the Predictors They Induce — arXiv:2609.20058},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.20058},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fb066441108a; sha256 95ed3c7469a1af4e5ed15fd4da66f9279a9397fbf0c5bebde37380acc66a167f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.20058. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Explanations of machine learning models are usually judged by criteria that are hard to compare. We propose a simpler test: if an explanation really describes how a model uses its features, it should be possible to rebuild the model's predictions from it. We turn each explanation into a predictor by reading each feature's effect and adding them up, and measure how well that predictor reproduces the model on unseen data. Nothing is fitted, so the score reflects the explanation itself. The test applies to any explanation that can be written as a function of the features; we demonstrate it on partial dependence plots (PDP), accumulated local effects (ALE), SHAP and LIME. We prove that summing partial dependence curves gives the best possible additive summary of a model when its features are independent, and that this fails when they are dependent. Across 13 real datasets and 9 synthetic designs and four model families, which method scores best depends entirely on feature dependence: where features are independent SHAP is slightly worse than PDP, exactly as the theory predicts; on dependent real data SHAP leads. Some widely used quality metrics even prefer a damaged explanation to an intact one. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-18 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.20058 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-10T03:48:18.312Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).