Enregistré dans:
Détails bibliographiques
Auteurs principaux: Kotsiopoulos, Konstandinos, Miroshnikov, Alexey, Filom, Khashayar, Kannan, Arjun Ravi
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:https://arxiv.org/abs/2303.10216
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866917644177768448
author Kotsiopoulos, Konstandinos
Miroshnikov, Alexey
Filom, Khashayar
Kannan, Arjun Ravi
author_facet Kotsiopoulos, Konstandinos
Miroshnikov, Alexey
Filom, Khashayar
Kannan, Arjun Ravi
contents In recent years, many Machine Learning (ML) explanation techniques have been designed using ideas from cooperative game theory. These game-theoretic explainers suffer from high complexity, hindering their exact computation in practical settings. In our work, we focus on a wide class of linear game values, as well as coalitional values, for the marginal game based on a given ML model and predictor vector. By viewing these explainers as expectations over appropriate sample spaces, we design a novel Monte Carlo sampling algorithm that estimates them at a reduced complexity that depends linearly on the size of the background dataset. We set up a rigorous framework for the statistical analysis and obtain error bounds for our sampling methods. The advantage of this approach is that it is fast, easily implementable, and model-agnostic. Furthermore, it has similar statistical accuracy as other known estimation techniques that are more complex and model-specific. We provide rigorous proofs of statistical convergence, as well as numerical experiments whose results agree with our theoretical findings.
format Preprint
id arxiv_https___arxiv_org_abs_2303_10216
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Approximation of group explainers with coalition structure using Monte Carlo sampling on the product space of coalitions and features
Kotsiopoulos, Konstandinos
Miroshnikov, Alexey
Filom, Khashayar
Kannan, Arjun Ravi
Machine Learning
Probability
In recent years, many Machine Learning (ML) explanation techniques have been designed using ideas from cooperative game theory. These game-theoretic explainers suffer from high complexity, hindering their exact computation in practical settings. In our work, we focus on a wide class of linear game values, as well as coalitional values, for the marginal game based on a given ML model and predictor vector. By viewing these explainers as expectations over appropriate sample spaces, we design a novel Monte Carlo sampling algorithm that estimates them at a reduced complexity that depends linearly on the size of the background dataset. We set up a rigorous framework for the statistical analysis and obtain error bounds for our sampling methods. The advantage of this approach is that it is fast, easily implementable, and model-agnostic. Furthermore, it has similar statistical accuracy as other known estimation techniques that are more complex and model-specific. We provide rigorous proofs of statistical convergence, as well as numerical experiments whose results agree with our theoretical findings.
title Approximation of group explainers with coalition structure using Monte Carlo sampling on the product space of coalitions and features
topic Machine Learning
Probability
url https://arxiv.org/abs/2303.10216