文档介绍：WASHINGTON UNIVERSITY
SEVER INSTITUTE OF TECHNOLOGY
DEPARTMENT PUTER SCIENCE AND ENGINEERING
COMPETENT PROGRAM EVOLUTION
by
Moshe Looks, ., .
Prepared under the direction of Professor R. P. Loui
A dissertation presented to the Sever Institute of
Washington University in partial fulfillment
of the requirements for the degree of
Doctor of Science
December 2006
Saint Louis, Missouri
WASHINGTON UNIVERSITY
SEVER INSTITUTE
SCHOOL OF ENGINEERING AND APPLIED SCIENCE
DEPARTMENT PUTER SCIENCE AND ENGINEERING
ABSTRACT
COMPETENT PROGRAM EVOLUTION
by Moshe Looks
ADVISOR: Professor R. P. Loui
December 2006
Saint Louis, Missouri
Heuristic optimization methods are adaptive when they sample problem solutions
based on knowledge of the search space gathered from past sampling. Recently, competent
evolutionary optimization methods have been developed that adapt via probabilistic mod-
eling of the search space. However, their effectiveness requires the existence of pact
problem position in terms of prespecified solution parameters.
How can we use these techniques to effectively and reliably solve program learn-
ing problems, given that program spaces will rarely pact positions? One
method is to manually build a problem-specific representation that is more tractable than
the general space. But can this process be automated? My thesis is that the properties
of programs and program spaces can be leveraged as inductive bias to reduce
the burden of manual representation-building, leading petent program
evolution.
The central contributions of this dissertation are a synthesis of the requirements
petent program evolution, and the design of a procedure, meta-optimizing seman-
tic evolutionary search (MOSES), that meets these requirements. In support of my thesis,
experimental results are provided to analyze and verify the effectiveness of MOSES, demon-
strating scalability and real-world applicability.
copyright by
Moshe Look