Why least squares and maximum entropy? An axiomatic approach to inference for linear inverse problems. (English) Zbl 0753.62003

Summary: An attempt is made to determine the logically consistent rules for selecting a vector from any feasible set defined by linear constraints, when either all \(n\)-vectors or those with positive components or the probability vectors are permissible. Some basic postulates are satisfied if and only if the selection rule is to minimize a certain function which, if a “ prior guess” is available, is a measure of distance from the prior guess. Two further natural postulates restrict the permissible distances to the author’s \(f\)-divergences and Bregman’s divergences [L. M. Bregman, USSR Comput. Math. Math. Phys. 7, No. 3, 200–217 (1969); translation from Zh. Vychisl. Mat. Mat. Fiz. 7, 620–631 (1967; Zbl 0186.23807)], respectively.
As corollaries, axiomatic characterizations of the methods of least squares and minimum discrimination information are arrived at. Alternatively, the latter are also characterized by a postulate of composition consistency. As a special case, a derivation of the method of maximum entropy from a small set of natural axioms is obtained.


62B10 Statistical aspects of information-theoretic topics
62A01 Foundations and philosophical topics in statistics
94A08 Image processing (compression, reconstruction, etc.) in information and communication theory


Zbl 0186.23807
Full Text: DOI