Computer Science > Machine Learning
This paper has been withdrawn by Vignesh Viswanathan
[Submitted on 8 Sep 2021 (v1), last revised 16 Feb 2024 (this version, v7)]
Title:Model Explanations via the Axiomatic Causal Lens
No PDF available, click to view other formatsAbstract:Explaining the decisions of black-box models is a central theme in the study of trustworthy ML. Numerous measures have been proposed in the literature; however, none of them take an axiomatic approach to causal explainability. In this work, we propose three explanation measures which aggregate the set of all but-for causes -- a necessary and sufficient explanation -- into feature importance weights. Our first measure is a natural adaptation of Chockler and Halpern's notion of causal responsibility, whereas the other two correspond to existing game-theoretic influence measures. We present an axiomatic treatment for our proposed indices, showing that they can be uniquely characterized by a set of desirable properties. We also extend our approach to derive a new method to compute the Shapley-Shubik and Banzhaf indices for black-box model explanations. Finally, we analyze and compare the necessity and sufficiency of all our proposed explanation measures in practice using the Adult-Income dataset. Thus, our work is the first to formally bridge the gap between model explanations, game-theoretic influence, and causal analysis.
Submission history
From: Vignesh Viswanathan [view email][v1] Wed, 8 Sep 2021 19:33:52 UTC (334 KB)
[v2] Fri, 17 Sep 2021 14:17:59 UTC (330 KB)
[v3] Mon, 31 Jan 2022 23:50:48 UTC (367 KB)
[v4] Mon, 11 Sep 2023 19:33:45 UTC (322 KB)
[v5] Wed, 27 Sep 2023 20:17:38 UTC (1 KB) (withdrawn)
[v6] Wed, 4 Oct 2023 20:36:32 UTC (367 KB)
[v7] Fri, 16 Feb 2024 00:16:03 UTC (1 KB) (withdrawn)
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
IArxiv Recommender
(What is IArxiv?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.