Soraya Rana’s research while affiliated with Colorado State University and other places

Parallel Genetic Algorithms have often been reported to yield better performance than Genetic Algorithms which use a single large panmictic population. In the case of the Island Model Genetic Algorithm, it has been informally argued that having multiple subpopulations helps to preserve genetic diversity, since each island can potentially follow a different search trajectory through the search space. It is also possible that since linearly separable problems are often used to test Genetic Algorithms, that Island models may simply be particularly well suited to exploiting the separable nature of the test problems. We explore this possibility by using the infinite population models of simple genetic algorithms to study how Island Models can track multiple search trajectories. We also introduce a simple model for better understanding when Island Model Genetic Algorithms may have an advantage when processing linearly separable problems.

In this paper, we examine the effects of noise on both local search and genetic search. Understanding the potential effects of noise on a search space may explain why some search techniques fail and why others succeed in the presence of noise. We discuss two effects that are the result of adding noise to a search space: the annealing of peaks in the search space and the introduction of false local optima.

When a function is mapped onto a bit representation, the structure of the fitness landscape can change dramatically. Techniques such as Delta Coding have tried to dynamically adapt the representation during the search process in hopes of making the problem easier for a genetic algorithm to solve.

We compare several techniques for scheduling shipment of customer orders for the Coors Brewing warehouse and production line. The goal is to minimize time at dock for trucks and railcars while also minimizing inventory. The techniques include a genetic algorithm, local search operators, heuristic rules, systematic search and hybrid approaches. Initial results show a hybrid genetic algorithm to be superior to the other methods. The evaluation function is a fast approximate form of a warehouse simulation. We also assess the sensitivity of the search algorithms to noise in an approximate evaluation function using a more detailed (and costly) simulation.

MAXSAT problems are notoriously difficult for genetic algorithms to solve. NKlandscapes are often used as test problems of scalable difficulty for genetic algorithms. In this paper we exploit the similar structure of the two problems to create an encompassing class of problems called embedded landscapes. Then we use Walsh analysis to explore the nonlinear bit interactions of these important test functions. We show that by applying Walsh analysis to embedded landscapes, several important summary statistics can be generated in polynomial time. We then use these techniques to discuss the statistical "shape" of both MAXSAT and NKlandscapes. 1 INTRODUCTION MAXSAT problems are notoriously difficult for genetic algorithms to solve. Even relatively old algorithms such as Davis-Putnam [Davis and Putnam, 1960] which are deterministic and exact are orders of magnitude faster than GAs. Understanding what makes MAXSAT so difficult for GAs gives us important clues about mechanisms of...

The Coors warehouse scheduling problem involves finding a permutation of customer orders that minimizes the average time that customers' orders spend at the loading docks while at the same time minimizing the running average inventory. Search based solutions require fast objective functions. Thus, a fast low-resolution simulation is used as an objective function. A slower high-resolution simulation is used to validate solutions. We compare the performance of a constructive scheduling algorithm to a genetic algorithm and local search approach. The constructive algorithm is based on a heuristic built specifically for this application. We also tested a hybrid of the genetic algorithm and local search approaches by initializing the search using the domain-specific heuristic. This hybrid genetic algorithm was able to find the best solutions when evaluated by the high-resolution simulation. Finally, we consider the effect of using the high-resolution simulation to filter a set ...

Parallel Genetic Algorithms have often been reported to yield better performance than Genetic Algorithms which use a single large panmictic population. In the case of the Island Model genetic algorithm, it has been informally argued that having multiple subpopulations helps to preserve genetic diversity, since each island can potentially follow a different search trajectory through the search space. It is also possible that since linearly separable problems are often used to test Genetic Algorithms, that Island Models may simply be particularly well suited to exploiting the separable nature of the test problems. We explore this possibility by using the infinite population models of simple genetic algorithms to study how Island Models can track multiple search trajectories. We also introduce a simple model for better understanding when Island Model genetic algorithms may have an advantage when processing some test problems. We provide empirical results for both linearly separa...

NK-landscapes and kSAT problems have been proposed as potential test problem domains for Genetic Algorithms. We demonstrate that GAs have difficulty solving both kSAT and NK-landscape problems. The construction of random kSAT and NK-landscape problems are very similar, but the differences between kSAT and NK-landscape generation result in vastly different fitness landscapes. In this paper we introduce a parameterized model for the construction of test function generators. This model, called embedded landscapes, can be used to isolate the features of combinatorial optimization problems for more control during experimentation. We also show that common forms of embedded landscapes allow for a polynomial time Walsh analysis. This means we also can compose exact schema averages in polynomial time for schema up to order-K, where K is a constant. Yet, in the general case, this information does not allow one to infer the global optimum of a function unless the complexity classes P ...

We compare several techniques for scheduling shipment of customer orders for the Coors Brewing warehouse and production line. The goal is to minimize time at dock for trucks and railcars while also minimizing inventory. The techniques include a genetic algorithm, local search operators, heuristic rules, systematic search and hybrid approaches. Initial results show a hybrid genetic algorithm to be superior to the other methods. The evaluation function is a fast approximate form of a warehouse simulation. We also assess the sensitivity of the search algorithms to noise in an approximate evaluation function using a more detailed (and costly) simulation

Random Boolean Satisfiability function generators have recently been proposed as tools for studying genetic algorithm behavior. Yet MAXSAT problems exhibit extremely limited epistasis. Furthermore, all nonzero Walsh coefficients can be computed exactly for MAXSAT problems in polynomial time using only the clause information. This means the low order schema averages can be computed quickly and exactly for very large MAXSAT problems. But unless P=NP, this low order information cannot reliably lead to the global optimum, thus nontrivial MAXSAT problems must be deceptive.