Explaining Multi-Criteria Decision Aiding Models with an Extended Shapley Value

The capability to explain the result of aggregation models to decision makers is key to reinforcing user trust. In practice, Multi-Criteria Decision Aiding models are often organized in a hierarchical way, based on a tree of criteria. We present an explanation approach usable with any hierarchical multicriteria model, based on an influence index of each attribute on the decision. A set of desirable axioms are defined. We show that there is a unique index fulfilling these axioms. This new index is an extension of the Shapley value on trees. An efficient rewriting of this index, drastically reducing the computation time, is obtained. Finally, the use of the new index is illustrated on an example.

[1]  David A. Klein,et al.  Decision-Analytic Intelligent Systems: Automated Explanation and Knowledge Acquisition , 1994 .

[2]  Frank Harary,et al.  Graph Theory , 2016 .

[3]  Scott Lundberg,et al.  A Unified Approach to Interpreting Model Predictions , 2017, NIPS.

[4]  Christophe Labreuche,et al.  Explaining robust additive utility models by sequences of preference swaps , 2015, Theory and Decision.

[5]  Isabelle Alvarez,et al.  Explaining the Result of a Decision Tree to the End-User , 2004, ECAI.

[6]  Jacky Montmain,et al.  Elucidation and Decisional Risk in a Multi Criteria Decision based on a Choquet Integral Aggregation—A Cybernetic Framework , 2005 .

[7]  Carlos José Pereira de Lucena,et al.  Pattern-based Explanation for Automated Decisions , 2014, ECAI.

[8]  Craig Boutilier,et al.  Constraint-based optimization and utility elicitation using the minimax decision criterion , 2006, Artif. Intell..

[9]  Anthony Hunter,et al.  On the measure of conflicts: Shapley Inconsistency Values , 2010, Artif. Intell..

[10]  Carlos Guestrin,et al.  "Why Should I Trust You?": Explaining the Predictions of Any Classifier , 2016, ArXiv.

[11]  Srdjan Vesic,et al.  Measuring the Intensity of Attacks in Argumentation Graphs with Shapley Value , 2017, IJCAI.

[12]  Michel Grabisch,et al.  Alternative Representations of Discrete Fuzzy Measures for Decision Making , 1997, Int. J. Uncertain. Fuzziness Knowl. Based Syst..

[13]  M. Grabisch.  The application of fuzzy integrals in multicriteria decision making , 1996 .

[14]  G. Owen.  VALUES OF GAMES WITH A PRIORI UNIONS , 1977 .

[15]  Johanna D. Moore,et al.  Generating and evaluating evaluative arguments , 2006, Artif. Intell..

[16]  Nils J. Nilsson,et al.  Artificial Intelligence , 1974, IFIP Congress.

[17]  G. A. Miller.  THE PSYCHOLOGICAL REVIEW THE MAGICAL NUMBER SEVEN, PLUS OR MINUS TWO: SOME LIMITS ON OUR CAPACITY FOR PROCESSING INFORMATION 1 , 1956 .

[18]  Wietske Visser,et al.  Explaining Qualitative Preference Models , 2012 .

[19]  G. Choquet.  Theory of capacities , 1954 .