Learning Good Variable Orderings

نویسنده

  • Paula Sturdy
چکیده

Variable ordering heuristics are used to reduce the cost of searching for a solution to a constraint satisfaction problem. On real problems that have non-binary and non-uniform constraints it is harder to make the optimal choice of variable ordering because surprisingly little is known about when and why variable ordering heuristics perform well. In an attempt to address this problem we present initial problem-specific investigations in variable orderings and suggest a method for determining good orderings using the cost of search.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

VARIABLE ORDERINGS FOR OBDDs, SIMULATED ANNEALING, AND THE HIDDEN WEIGHTED BIT FUNCTION

Ordered binary decision diagrams (OBDDs) are an eecient graph representation for Boolean functions, if good variable orderings are used. Variable order-ings are computed by heuristic algorithms and improved with the simulated annealing approach. New neighborhoods for simulated annealing algorithms are presented. It is investigated how the OBDD size may change in one iteration step for the diier...

متن کامل

Variable ordering for ordered binary decision diagrams by a divide-and-conquer approach

An efficient variable ordering strategy for ordered binary decision diagrams (OBDD) based on interleaving the compacted clusters is proposed in this paper. The novelty of this method is to apply the divide-and-conquer approach to find a good variable ordering efficiently for circuits with a large number of I/ Os. First, a given circuit is partitioned into a number of clusters according to the c...

متن کامل

Stochastic Local Algorithms for Learning Belief Networks: Searching in the Space of the Orderings

An important type of methods for learning belief networks from data are those based on the use of a scoring metric, to evaluate the fitness of any given candidate network to the data base, and a search procedure to explore the set of candidate networks. In this paper we propose a new method that carries out the search not in the space of directed acyclic graphs but in the space of the orderings...

متن کامل

Karlin’s Basic Composition Theorems and Stochastic Orderings

Suppose λ,x,ζ traverse the ordered sets Λ, X and Z, respectively and consider the functions f(λ,x,ζ) and g(λ,ζ) satisfying the following conditions, (a) f(λ,x,ζ) > 0 and f is TP2 in each pairs of variables when the third variable is held fixed and (b) g(λ,ζ) is TP2. Then the function h(λ,x) =∫Z f(λ,x,ζ)g(λ,ζ)dµ(ζ), defined on Λ×X is TP2 in (λ,x). The aim of this note is to use a new stochast...

متن کامل

Optimal ordered binary decision diagrams for read-once formulas

In many applications like verification or combinatorial optimization, OBDDs (ordered binary decision diagrams) are used as a representation or data structure for Boolean functions. Efficient algorithms exist for the important operations on OBDDs, and many functions can be represented in reasonable size if a good variable ordering is chosen. In general, it is NP-hard to compute optimal or near-o...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2003