A measure-theoretic analysis of stochastic optimization

This paper proposes a measure-theoretic framework to study iterative stochastic optimizers that provides theoretical tools to explore how the optimization methods may be improved. Within this framework, optimizers form a closed, convex subset of a normed vector space, implying the existence of a distance metric between any two optimizers and a meaningful and computable spectrum of new optimizers between them. It is shown how the formalism applies to evolutionary algorithms in general. The analytic property of continuity is studied in the context of genetic algorithms, revealing the conditions under which approximations such as meta-modeling or surrogate methods may be effective. These results demonstrate the power of the proposed analytic framework, which can be used to propose and analyze new techniques such as controlled convex combinations of optimizers, meta-optimization of algorithm parameters, and more.

[1]  Michael D. Vose,et al.  The simple genetic algorithm - foundations and theory , 1999, Complex adaptive systems.

[2]  David H. Wolpert,et al.  No free lunch theorems for optimization , 1997, IEEE Trans. Evol. Comput..

[3]  Olivier Teytaud,et al.  Continuous lunches are free! , 2007, GECCO '07.

[4]  Kenneth O. Stanley,et al.  Abandoning Objectives: Evolution Through the Search for Novelty Alone , 2011, Evolutionary Computation.

[5]  Risto Miikkulainen,et al.  Real-space evolutionary annealing , 2011, GECCO '11.

[6]  Patrick Billingsley,et al.  Probability and Measure. , 1986 .

[7]  Tom Schaul,et al.  Curiosity-driven optimization , 2011, 2011 IEEE Congress of Evolutionary Computation (CEC).

[8]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[9]  Patrick D. Surry,et al.  Fundamental Limitations on Search Algorithms: Evolutionary Computing in Perspective , 1995, Computer Science Today.

[10]  C. D. Gelatt,et al.  Optimization by Simulated Annealing , 1983, Science.

[11]  P. Spreij Probability and Measure , 1996 .

[12]  Yaochu Jin,et al.  Surrogate-assisted evolutionary computation: Recent advances and future challenges , 2011, Swarm Evol. Comput..

[13]  Andy J. Keane,et al.  Metamodeling Techniques For Evolutionary Optimization of Computationally Expensive Problems: Promises and Limitations , 1999, GECCO.

[14]  L. D. Whitley,et al.  The No Free Lunch and problem description length , 2001 .

[15]  M. Vose The Simple Genetic Algorithm , 1999 .

[16]  Michael D. Vose,et al.  Random Heuristic Search , 1999, Theor. Comput. Sci..