Title :

The Tractability of SHAP-Score-Based Explanations over Deterministic and Decomposable Boolean Circuits

Author(s) :

Arenas, Marcelo [Auteur]
Millennium Institute for Foundational Research on Data [IMFD]
Pontificia Universidad Católica de Chile [UC]
Barceló, Pablo [Auteur]
Pontificia Universidad Católica de Chile [UC]
Millennium Institute for Foundational Research on Data [IMFD]
Bertossi, Leopoldo [Auteur]
Universidad Adolfo Ibáñez [Santiago]
Millennium Institute for Foundational Research on Data [IMFD]
Monet, Mikaël [Auteur]
Linking Dynamic Data [LINKS]

Conference title :

AAAI 2021 - 35th Conference on Artificial Intelligence

City :

Virtual

Country :

France

Start date of the conference :

2021-02-02

HAL domain(s) :

Informatique [cs]

English abstract : [en]

Scores based on Shapley values are widely used for providing explanations to classification results over machine learning models. A prime example of this is the influential SHAPscore, a version of the Shapley value that ...
Show more >Scores based on Shapley values are widely used for providing explanations to classification results over machine learning models. A prime example of this is the influential SHAPscore, a version of the Shapley value that can help explain the result of a learned model on a specific entity by assigning a score to every feature. While in general computing Shapley values is a computationally intractable problem, it has recently been claimed that the SHAP-score can be computed in polynomial time over the class of decision trees. In this paper, we provide a proof of a stronger result over Boolean models: the SHAP-score can be computed in polynomial time over deterministic and decomposable Boolean circuits. Such circuits, also known as tractable Boolean circuits, generalize a wide range of Boolean circuits and binary decision diagrams classes, including binary decision trees, Ordered Binary Decision Diagrams (OBDDs) and Free Binary Decision Diagrams (FBDDs). We also establish the computational limits of the notion of SHAP-score by observing that, under a mild condition, computing it over a class of Boolean models is always polynomially as hard as the model counting problem for that class. This implies that both determinism and decomposability are essential properties for the circuits that we consider, as removing one or the other renders the problem of computing the SHAP-score intractable (namely, #P-hard).Show less >

Language :

Anglais

Peer reviewed article :

Oui

Audience :

Internationale

Popular science :

Non

Collections :

• Centre de Recherche en Informatique, Signal et Automatique de Lille (CRIStAL) - UMR 9189

Source :

Harvested from HAL