K-Optimal Design via Semidefinite Programming and Entropy Optimization

In this paper, we consider the problem of optimal design of experiments. A two-step inference strategy is proposed. The first step consists in minimizing the condition number of the so-called information matrix. This step can be turned into a semidefinite programming problem. The second step is more classical, and it entails the minimization of a convex integral functional under linear constraints. This step is formulated in some infinite-dimensional space and is solved by means of a dual approach. Numerical simulations will show the relevance of our approach.

[1]  Jonathan M. Borwein,et al.  Partially finite convex programming, Part I: Quasi relative interiors and duality theory , 1992, Math. Program..

[2]  F. Pukelsheim.  Optimal Design of Experiments , 1993 .

[3]  Andrej Pázman,et al.  Foundations of Optimum Experimental Design , 1986 .

[4]  Ting Kei Pong,et al.  Methods for Optimal Experimental Designs ∗ , 2012 .

[5]  R. Tyrrell Rockafellar,et al.  Convex Integral Functionals and Duality , 1971 .

[6]  K. Chaloner,et al.  Bayesian Experimental Design: A Review , 1995 .

[7]  D'avid Papp,et al.  Optimal Designs for Rational Function Regression , 2010, 1009.1444.

[8]  R. Rockafellar.  Integrals which are convex functionals. II , 1968 .

[9]  Jane J. Ye,et al.  Optimizing Condition Numbers , 2009, SIAM J. Optim..

[10]  R. Rockafellar,et al.  Integral functionals, normal integrands and measurable selections , 1976 .

[11]  Radoslav Harman,et al.  Approximate D-optimal designs of experiments on the convex hull of a finite set of information matrices , 2009 .

[12]  Holger Dette,et al.  Improving updating rules in multiplicative algorithms for computing D-optimal designs , 2008, Comput. Stat. Data Anal..

[13]  Xiaojun Chen,et al.  Minimizing the Condition Number of a Gram Matrix , 2011, SIAM J. Optim..

[14]  J. I.  The Design of Experiments , 1936, Nature.

[15]  Yaming Yu.  Monotonic convergence of a general algorithm for computing optimal designs , 2009, 0905.2646.

[16]  Julie Zhou,et al.  D-optimal minimax regression designs on discrete design space , 2008 .

[17]  J. Hiriart-Urruty,et al.  Convex analysis and minimization algorithms , 1993 .

[18]  R. A. Leibler,et al.  On Information and Sufficiency , 1951 .

[19]  R. Rockafellar.  Convex Analysis: (pms-28) , 1970 .

[20]  丸山 徹.  Convex Analysisの二,三の進展について , 1977 .

[21]  C. Zălinescu.  Convex analysis in general vector spaces , 2002 .

[22]  C. M. Shetty,et al.  Nonlinear Programming - Theory and Algorithms, Second Edition , 1993 .

[23]  K. Schittkowski,et al.  NONLINEAR PROGRAMMING , 2022 .

[24]  Mokhtar S. Bazaraa,et al.  Nonlinear Programming: Theory and Algorithms , 1993 .

[25]  Jonathan M. Borwein,et al.  On the convergence of moment problems , 1991 .

[26]  Ting Kei Pong,et al.  Minimizing Condition Number via Convex Programming , 2011, SIAM J. Matrix Anal. Appl..

[27]  Valerii Fedorov,et al.  Design of Experiments in Statistics , 2000 .

[28]  Bastian Goldlücke,et al.  Variational Analysis , 2014, Computer Vision, A Reference Guide.

[29]  Jane J. Ye,et al.  Minimizing the Condition Number to Construct Design Points for Polynomial Regression Models , 2013, SIAM J. Optim..

[30]  J. Borwein,et al.  Duality relationships for entropy-like minimization problems , 1991 .

[31]  R. Martín-Martín,et al.  Multiplicative algorithms for computing optimum designs , 2009 .