Measuring the interactions among variables of functions over the unit hypercube

DOI10.1016/J.JMAA.2011.02.040MaRDI QIDQ534728zbMATH OpenOpenAlexFDO

Authors Jean-Luc Marichal, Pierre Mathonet

Publication date 10 May 2011

Published in Journal of Mathematical Analysis and Applications (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/0912.1547

zbMATH Keywords

aggregation functionBanzhaf interaction indexcooperative (fuzzy) gameleast squares approximationmultilinear polynomial

Mathematics Subject Classification ID

Cooperative games (91A12) Games in extensive form (91A18)

Abstract: By considering a least squares approximation of a given square integrable function $fcolon[0,1{]}^{n}oR$ by a multilinear polynomial of a specified degree, we define an index which measures the overall interaction among variables of $f$. This definition extends the concept of Banzhaf interaction index introduced in cooperative game theory. Our approach is partly inspired from multilinear regression analysis, where interactions among the independent variables are taken into consideration. We show that this interaction index has appealing properties which naturally generalize the properties of the Banzhaf interaction index. In particular, we interpret this index as an expected value of the difference quotients of $f$ or, under certain natural conditions on $f$, as an expected value of the derivatives of $f$. These interpretations show a strong analogy between the introduced interaction index and the overall importance index defined by Grabisch and Labreuche [7]. Finally, we discuss a few applications of the interaction index.

Review text

By considering a least squares approximation of a given square integrable function over the unit hypercube and following the interaction in cooperative (fuzzy) game theory, in analysis and in statistics, the authors generalize the Banzhaf interaction index and prove some corresponding properties like linearity, continuity and symmetry. Certain applications in aggregation function theory together with some implications of the interaction index to compare different functions through the interaction degrees of their variables and to measure the quality of multilinear approximations complete the paper under significant references.

Recommendations

• Influence and interaction indexes for pseudo-Boolean functions: a unified least squares approach
• Modeling interaction phenomena using fuzzy measures: On the notions of interaction and independence
• An axiomatic approach to the measurement of the amount of interaction among criteria or players
• Interaction indices revisited
• Approximations of Lovász extensions and their induced interaction index

Cites work

• Absorbent tuples of aggregation operators
• Aggregation functions.
• An axiomatic approach to the concept of interaction among players in cooperative games
• An introduction to copulas.
• Approximations of Lovász extensions and their induced interaction index
• Approximations of pseudo-Boolean functions; applications to game theory
• Axiomatic characterizations of generalized values
• Axiomatic characterizations of probabilistic and cardinal-probabilistic interaction indices
• Cooperative Fuzzy Games
• Equivalent representations of set functions
• Fuzzy measures and integrals. Theory and applications
• HOW TO IMPROVE ACTS: AN ALTERNATIVE REPRESENTATION OF THE IMPORTANCE OF CRITERIA IN MCDM
• scientific article; zbMATH DE number 4106654 (Why is no real title available?)
• scientific article; zbMATH DE number 3249560 (Why is no real title available?)
• scientific article; zbMATH DE number 3078997 (Why is no real title available?)
• Mathematical Properties of the Banzhaf Power Index
• The influence of variables in product spaces
• The influence of variables on pseudo-Boolean functions with applications to game theory and multicriteria decision making

Cited in

(2)

• Derivative of functions over lattices as a basis for the notion of interaction between attributes
• scientific article; zbMATH DE number 1390004 (Why is no real title available?)

This page was built for publication: Measuring the interactions among variables of functions over the unit hypercube

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q534728)