Maximum entropy algorithm with inexact upper entropy bound based on Fup basis functions with compact support

Authors:
Hrvoje Gotovac;Bla Gotovac
Affiliations:
Department of Land and Water Resources Engineering, KTH, Brinellvagen 32,10044 Stockholm, Sweden and Department of Civil and Architectural Engineering, University of Split, Matice hrvatske 15, 210 ...;Department of Civil and Architectural Engineering, University of Split, Matice hrvatske 15, 21000 Split, Croatia
Venue:
Journal of Computational Physics
Year:
2009

Citing 4
Cited 0

A maximum entropy approach to natural language processing

Computational Linguistics
A practical computational framework for the multidimensional moment-constrained maximum entropy principle

Journal of Computational Physics
Atomic functions: Generalization to the multivariable case and promising applications

Cybernetics and Systems Analysis
The multidimensional moment-constrained maximum entropy problem: A BFGS algorithm with constraint scaling

Journal of Computational Physics

Quantified Score

Hi-index	31.45

Visualization

Abstract

The maximum entropy (MaxEnt) principle is a versatile tool for statistical inference of the probability density function (pdf) from its moments as a least-biased estimation among all other possible pdf's. It maximizes Shannon entropy, satisfying the moment constraints. Thus, the MaxEnt algorithm transforms the original constrained optimization problem to the unconstrained dual optimization problem using Lagrangian multipliers. The Classic Moment Problem (CMP) uses algebraic power moments, causing typical conventional numerical methods to fail for higher-order moments (m5-10) due to different sensitivities of Lagrangian multipliers and unbalanced nonlinearities. Classic MaxEnt algorithms overcome these difficulties by using orthogonal polynomials, which enable roughly the same sensitivity for all Lagrangian multipliers. In this paper, we employ an idea based on different principles, using Fup"n basis functions with compact support, which can exactly describe algebraic polynomials, but only if the Fup order-n is greater than or equal to the polynomial's order. Our algorithm solves the CMP with respect to the moments of only low order Fup"2 basis functions, finding a Fup"2 optimal pdf with better balanced Lagrangian multipliers. The algorithm is numerically very efficient due to localized properties of Fup"2 basis functions implying a weaker dependence between Lagrangian multipliers and faster convergence. Only consequences are an iterative scheme of the algorithm where power moments are a sum of Fup"2 and residual moments and an inexact entropy upper bound. However, due to small residual moments, the algorithm converges very quickly as demonstrated on two continuous pdf examples - the beta distribution and a bi-modal pdf, and two discontinuous pdf examples - the step and double Dirac pdf. Finally, these pdf examples present that Fup MaxEnt algorithm yields smaller entropy value than classic MaxEnt algorithm, but differences are very small for all practical engineering purposes.