Results 1  10
of
44
PopulationBased Incremental Learning: A Method for Integrating Genetic Search Based Function Optimization and Competitive Learning
, 1994
"... Genetic algorithms (GAs) are biologically motivated adaptive systems which have been used, with varying degrees of success, for function optimization. In this study, an abstraction of the basic genetic algorithm, the Equilibrium Genetic Algorithm (EGA), and the GA in turn, are reconsidered within th ..."
Abstract

Cited by 335 (11 self)
 Add to MetaCart
(Show Context)
Genetic algorithms (GAs) are biologically motivated adaptive systems which have been used, with varying degrees of success, for function optimization. In this study, an abstraction of the basic genetic algorithm, the Equilibrium Genetic Algorithm (EGA), and the GA in turn, are reconsidered within the framework of competitive learning. This new perspective reveals a number of different possibilities for performance improvements. This paper explores populationbased incremental learning (PBIL), a method of combining the mechanisms of a generational genetic algorithm with simple competitive learning. The combination of these two methods reveals a tool which is far simpler than a GA, and which outperforms a GA on large set of optimization problems in terms of both speed and accuracy. This paper presents an empirical analysis of where the proposed technique will outperform genetic algorithms, and describes a class of problems in which a genetic algorithm may be able to perform better. Extensions to this algorithm are discussed and analyzed. PBIL and extensions are compared with a standard GA on twelve problems, including standard numerical optimization functions, traditional GA test suite problems, and NPComplete problems.
Benchmarks for Basic Scheduling Problems
, 1989
"... In this paper, we propose 260 scheduling problems whose size is greater than that of the rare examples published. Such sizes correspond to real dimensions of industrial problems. ..."
Abstract

Cited by 203 (0 self)
 Add to MetaCart
In this paper, we propose 260 scheduling problems whose size is greater than that of the rare examples published. Such sizes correspond to real dimensions of industrial problems.
DESIGN, IMPLEMENTATION, AND EVALUATION OF THE CONSTRAINT LANGUAGE cc(FD)
 J. LOGIC PROGRAMMING 1994:19, 20:1679
, 1994
"... This paper describes the design, implementation, and applications of the constraint logic language cc(FD). cc(FD) is a declarative nondeterministic constraint logic language over finite domains based on the cc framework [33], an extension of the CLP scheme [21]. Its constraint solver includes (nonl ..."
Abstract

Cited by 183 (10 self)
 Add to MetaCart
This paper describes the design, implementation, and applications of the constraint logic language cc(FD). cc(FD) is a declarative nondeterministic constraint logic language over finite domains based on the cc framework [33], an extension of the CLP scheme [21]. Its constraint solver includes (nonlinear) arithmetic constraints over natural numbers which are approximated using domain and interval consistency. The main novelty of cc(FD) is the inclusion of a number of generalpurpose combinators, in particular cardinality, constructive disjunction, and blocking implication, in conjunction with new constraint operations such as constraint entailment and generalization. These combinators significantly improve the operational expressiveness, extensibility, and flexibility of CLP languages and allow issues such as the definition of nonprimitive constraints and disjunctions to be tackled at the language level. The implementation of cc(FD) (about 40,000 lines of C) includes a WAMbased engine [44], optimal arcconsistency algorithms based on AC5 [40], and incremental implementation of the combinators. Results on numerous problems, including scheduling, resource allocation, sequencing, packing, and hamiltonian paths are reported and indicate that cc(FD) comes close to procedural languages on a number of combinatorial problems. In addition, a small cc(FD) program was able to find the optimal solution and prove optimality to a famous 10/10 disjunctive scheduling problem [29], which was left open for more than 20 years and finally solved in 1986.
Improved CLP Scheduling with Task Intervals
, 1994
"... In this paper we present a new technique that can be used to improve performance of job scheduling with a constraint programming language. We show how, by focusing on some special sets of tasks, one can bring CLP in the same range of efficiency as traditional OR algorithms on a classical benchmark ( ..."
Abstract

Cited by 98 (6 self)
 Add to MetaCart
(Show Context)
In this paper we present a new technique that can be used to improve performance of job scheduling with a constraint programming language. We show how, by focusing on some special sets of tasks, one can bring CLP in the same range of efficiency as traditional OR algorithms on a classical benchmark (MT10 [MT63]), thus making CLP both a flexible and an efficient technique for such combinatorial problems. We then present our programming methodology which we have successfully used on many problems, and draw conclusions on what features constraint programming languages should offer to allow its use. 1. Introduction Reallife scheduling problems are often the composition of various wellidentified hard problems. In the previous years, we have worked on applications such as tasktechnician assignments [CK92] or staff timetable scheduling [CGL93] and developed a methodology for solving such problems with an extensible constraint logic programming language. In both cases we have applied the s...
A Promising Genetic Algorithm Approach to JobShop Scheduling, Rescheduling, and OpenShop Scheduling Problems
 Proceedings of the Fifth International Conference on Genetic Algorithms
, 1993
"... The general jobshop scheduling problem is known to be extremely hard. We describe a GA approach which produces reasonably good results very quickly on standard benchmark jobshop scheduling problems, better than previous efforts using genetic algorithms for this task, and comparable to existing con ..."
Abstract

Cited by 89 (3 self)
 Add to MetaCart
(Show Context)
The general jobshop scheduling problem is known to be extremely hard. We describe a GA approach which produces reasonably good results very quickly on standard benchmark jobshop scheduling problems, better than previous efforts using genetic algorithms for this task, and comparable to existing conventional searchbased methods. The representation used is a variant of one known to work moderately well for the traveling salesman problem. It has the considerable merit that crossover will always produce legal schedules. A novel method for performance enhancement is examined based on dynamic sampling of the convergence rates in different parts of the genome. Our approach also promises to effectively address the openshop scheduling problem and the jobshop rescheduling problem. 1 INTRODUCTION The jobshop scheduling problem (JSSP) is a very important practical problem. Efficient methods of solving it can have major effects on profitability and product quality, but with the JSSP being amon...
An Empirical Comparison of Seven Iterative and Evolutionary Function Optimization Heuristics
, 1995
"... This report is a repository for the results obtained from a large scale empirical comparison of seven iterative and evolutionbased optimization heuristics. Twentyseven static optimization problems, spanning six sets of problem classes which are commonly explored in genetic algorithm literature, ..."
Abstract

Cited by 54 (8 self)
 Add to MetaCart
This report is a repository for the results obtained from a large scale empirical comparison of seven iterative and evolutionbased optimization heuristics. Twentyseven static optimization problems, spanning six sets of problem classes which are commonly explored in genetic algorithm literature, are examined. The problem sets include jobshop scheduling, traveling salesman, knapsack, binpacking, neural network weight optimization, and standard numerical optimization. The search spaces in these problems range from 2^368 to 2^2040. The results indicate that using genetic algorithms for the optimization of static functions does not yield a benefit, in terms of the final answer obtained, over simpler optimization heuristics. Descriptions of the algorithms tested and the encodings of the problems are described in detail for reproducibility.
Stochastic Hillclimbing as a Baseline Method for Evaluating Genetic Algorithms
, 1994
"... We investigate the effectiveness of stochastic hillclimbing as a baseline for evaluating the performance of genetic algorithms (GAs) as combinatorial function optimizers. In particular, we address four problems to which GAs have been applied in the literature: the maximum cut problem, Koza's 11 ..."
Abstract

Cited by 51 (0 self)
 Add to MetaCart
(Show Context)
We investigate the effectiveness of stochastic hillclimbing as a baseline for evaluating the performance of genetic algorithms (GAs) as combinatorial function optimizers. In particular, we address four problems to which GAs have been applied in the literature: the maximum cut problem, Koza's 11multiplexer problem, MDAP (the Multiprocessor Document Allocation Problem), and the jobshop problem. We demonstrate that simple stochastic hillclimbing methods are able to achieve results comparable or superior to those obtained by the GAs designed to address these four problems. We further illustrate, in the case of the jobshop problem, how insights obtained in the formulation of a stochastic hillclimbing algorithm can lead to improvements in the encoding used by a GA. Department of Computer Science, University of California at Berkeley. Supported by a NASA Graduate Fellowship. This paper was written while the author was a visiting researcher at the Ecole Normale Sup'erieurerue d'Ulm, Group...
Fast Probabilistic Modeling for Combinatorial Optimization
 AAAI98
, 1998
"... Probabilistic models have recently been utilized for the optimization of large combinatorial search problems. However, complex probabilistic models that attempt to capture interparameter dependencies can have prohibitive computational costs. The algorithm presented in this paper, termed COMIT, provi ..."
Abstract

Cited by 39 (1 self)
 Add to MetaCart
Probabilistic models have recently been utilized for the optimization of large combinatorial search problems. However, complex probabilistic models that attempt to capture interparameter dependencies can have prohibitive computational costs. The algorithm presented in this paper, termed COMIT, provides a method for using probabilistic models in conjunction with fast search techniques. We show how COMIT can be used with two very different fast search algorithms: hillclimbing and Populationbased incremental learning (PBIL). The resulting algorithms maintain many of the benefits of probabilistic modeling, with far less computational expense. Extensive empirical results are provided; COMIT has been successfully applied to jobshop scheduling, traveling salesman, and knapsack problems. This paper also presents a review of probabilistic modeling for combinatorial optimization.
Finite Domain Constraint Programming in Oz  A Tutorial
 PROGRAMMING SYSTEMS LAB, GERMAN RESEARCH CENTER FOR ARTIFICIAL INTELLIGENCE (DFKI
, 1998
"... This document introduces constraint programming in Oz. We restrict our attention to combinatorial problems that can be stated with variables ranging over finite sets of nonnegative integers. Problems in this class range from puzzles to real world applications as diverse as scheduling, ware house all ..."
Abstract

Cited by 37 (3 self)
 Add to MetaCart
This document introduces constraint programming in Oz. We restrict our attention to combinatorial problems that can be stated with variables ranging over finite sets of nonnegative integers. Problems in this class range from puzzles to real world applications as diverse as scheduling, ware house allocation, configuration and placement.
Population Based Incremental Learning: A Method for Integrating Genetic Search Based Function Optimization and Competitve Learning
, 1994
"... Genetic algorithms (GAs) are biologically motivated adaptive systems which have been used, with varying degrees of success, for function optimization. In this study, an abstraction of the basic genetic algorithm, the Equilibrium Genetic Algorithm (EGA), and the GA in turn, are reconsidered within ..."
Abstract

Cited by 35 (0 self)
 Add to MetaCart
Genetic algorithms (GAs) are biologically motivated adaptive systems which have been used, with varying degrees of success, for function optimization. In this study, an abstraction of the basic genetic algorithm, the Equilibrium Genetic Algorithm (EGA), and the GA in turn, are reconsidered within the framework of competitive learning. This new perspective reveals a number of different possibilities for performance improvements. This paper explores population based incremental learning (PBIL), a method of combining the mechanisms of a generational genetic algorithm with simple competitive learning. The combination of these two methods reveals a tool which is far simpler than a GA, and which outperforms a GA on large set of optimization problems in terms of both speed and accuracy. This paper presents an empirical analysis of where the proposed technique will outperform genetic algorithms, and describes a class of problems in which a genetic algorithm may be able to perform b...