Automatic Construction of Gaussian-Based Edge Detectors Using Genetic Programming

The 0/1 knapsack problem is one of the most exhaustively studied NP-hard combinatorial optimization problems. Many different approaches have been taken to obtain an approximate solution to the problem in polynomial time. Here we consider the biobjective 0/1 knapsack problem. The contribution of this paper is to show that a genetic programming system can evolve a set of heuristics that can give solutions on the Pareto front for multiobjective combinatorial problems. The genetic programming (GP) system outlined here evolves a heuristic which decides whether or not to add an item to the knapsack in such a way that the final solution is one of the Pareto optimal solutions. Moreover, the Pareto front obtained from the GP system is comparable to the front obtained from other human-designed heuristics. We discuss the issue of the diversity of the obtained Pareto front and the application of strongly-typed GP as a means of obtaining better diversity.

Document classification tasks generally have sparse and high dimensional features. It is important to effectively extract features. In document classification tasks, there are some similarities existing in different categories or different datasets. It is possible that one document classification task does not have labelled training data. In order to obtain effective classifiers on this specific task, this paper proposes a Genetic Programming (GP) system using transductive transfer learning. The proposed GP system automatically extracts features from different source domains, and these GP extracted features are combined to form new classifiers being directly applied to a target domain. From experimental results, the proposed transductive transfer learning GP system can evolve features from source domains to effectively apply to target domains which are similar to the source domains.

Algorithms are needed in every aspect of parallel computing. Genetic Programming is an evolutionary technique for automating the design of algorithms through iterative steps of mutation and crossover operations on an initial population of randomly generated computer programs. This paper describes a novel parallel genetic programming (GP) system inspired by the symbiogenesis model of evolution, wherein new organisms are generated through the absorption of different life-forms in addition to the usual mutation and crossover operations. Different organisms are expressed in this GP system through multiple program representations. Two program representations considered in this paper are the procedural representation (PR) and the tree representation (TR). Populations of these representations evolve separately. Individuals in each population migrate to the other and participate in evolution via representation change algorithms. Parallelism is achieved through use of the AutoMap/AutoLink MPI library. The differences in the locality properties of the representations serve as a source of new ideas for creating the final algorithm.

Gaussian-based filtering techniques have been popularly applied to edge detection. However, how to effectively tune parameters of Gaussian filters and how to effectively combine different Gaussian filters are still open issues. In this study, a new genetic programming (GP) approach is proposed to automatically tune parameters of Gaussian filters and automatically combine different types of Gaussian filters to extract edge features. In general, it is time-consuming for GP to evolve edge detectors using a large training image dataset. To efficiently evolve edge detectors from a large training image dataset, we propose sampling techniques (randomly selecting training images) to evolve Gaussian-based edge detectors. A sampling technique only using part of a set of images obtains similar performance to the training data using all of these images. The evolved edge detectors from the proposed sampling technique perform better than the Gaussian gradient and rotation invariant surround suppression. Based on the analysis of GP evolving edge detectors, it is suggested that combining Gaussian filters should be based on different types of Gaussian filters, and the Gaussian gradient should be considered as a major filter in these combinations.

Genetic Programming (GP) is a form of nature-inspired computing, introduced over 30 years ago, with notable success in problems such as symbolic regression. However, there remains a lot of relatively unexploited potential for solving hard, real-world problems. There is consensus in the GP community that the lack of effective real-world benchmark problems negatively impacts the quality of research [4]. When a GP system is initialised, a number of parameters must be provided. The optimal setup configuration is often not known, due to the fact that many of the values are problem and domain specific, meaning the GP system is unable to produce satisfactory results. We believe that the implementation of continuous self-adaptation, along with the introduction of tunable and suitably difficult benchmark problems, will allow for the creation of more robust GP systems that are resilient to failure.

In this work, genetic programming systems are viewed as cooperative processes: individuals in the population cooperate to evolve a global solution. There are two basic forms of cooperation for distributed problem solving: task-sharing and result-sharing. We introduce two models that enable cooperation in genetic programming systems. The first model is based on task-sharing and the second one is based on result-sharing. We examine the effects of both forms of cooperation on the performance of genetic programming systems solving the insect locomotion problem. This article demostrates that cooperative genetic programming can be used to evolve several commonly observed gaits in insects.

<p>One of the greater issues in Genetic Programming (GP) is the computational effort required to run the evolution and discover a good solution. Phenomena such as program bloating (where genetic programs rapidly grow in size) can quickly exhaust available memory resources and slow down the evolutionary process, while the heavy cost of performing fitness evaluation can make problems which have a lot of available data very slow to solve. These issues may limit GP in some tasks it can appropriately be applied to, as well as inhibit its applications in time/space sensitive environments. In this thesis, we look at developing solutions to some of these issues in GP computational cost. First, we develop an algebraic program simplification method based on simple rules and hashing techniques, and use this method in conjunction with the standard GP on a variety of tasks. Our results suggest that program simplification can lead to a significant reduction in program size, while not significantly changing the effectiveness of the systems in finding solution programs. Secondly, we analyse the effects of program simplification on the internal GP "building blocks" to investigate whether simplification is a destructive or constructive force. Using two models for building blocks (numerical-nodes and the more complex fixed-depth subtree), we track building blocks through GP runs on a symbolic regression problem, both with and without using simplification. We find that the program simplification process can both disrupt and construct building blocks in the GP populations. However, GP systems using simplification appear to retain important building blocks, and the simplification process appears to lead to an increase in genetic diversity. These may help explain why using simplification does not reduce the effectiveness of GP systems in solving tasks. Lastly, we develop two methods of reducing the cost of fitness evaluation by reducing the number of node evaluations performed. The first method is elitism avoidance, which avoids re-evaluating programs which have been placed in the population using elitismreproduction. Thismethod reduces the CPU time for evolving solutions for six different GP tasks. The second method is a subtree caching mechanism which store fitness evaluations for subtrees in a cache so that they may be fetched when these subtrees are encountered in future fitness evaluations. Results suggest that using this mechanism can significantly reduce both the number of node evaluations and the overall CPU time used in evolving solutions, without reducing the fitness of the solutions produced.</p>

Genetic programming (GP) is applied to the design of fuzzy logic controllers (FLCs) for mobile robot path tracking. GP is applied to automatic discovery of full knowledge bases for use in fuzzy logic control applications. An extension to a rule learning GP system is presented that achieves this objective. In addition, GP is employed to handle selection of fuzzy set intersection operators (t-norms). The new GP system is applied to design a mobile robot path tracking controller and performance is shown to be comparable to that of a manually designed controller. GP was successfully applied to discover FLCs capable of steering a mobile robot to track straight-line paths in the plane. Instances of simultaneous evolution of membership functions and rules showed that GP was capable of evolving a FLC that demonstrated satisfactory responsiveness to various initial conditions while utilizing minimal human interface.

Programming languages provide a variety of mechanisms to associate names with values, and these mechanisms play a central role in programming practice. For example, they allow multiple references to the same storage location or function in different parts of a complex program. By contrast, the representations used in current genetic programming systems provide few if any naming mechanisms, and it is therefore generally not possible for evolved programs to use names in sophisticated ways. In this chapter we describe a new approach to names in genetic programming that is based on Holland’s concept of tags. We demonstrate the use of tag-based names, we describe some of the ways in which they may help to extend the power and reach of genetic programming systems, and we look at the ways that tag-based names are actually used in an evolved program that solves a robot navigation problem.Keywordsgenetic programmingmodularitynamestagsstack-based genetic programmingPush programming languagePushGP

The combination of local features in edge detection can generally improve detection performance. However, how to effectively combine different basic features remains an open issue and needs to be investigated. Multivariate density is a generalisation of the one-dimensional (univariate) distribution to higher dimensions. In order to effectively construct composite features with multivariate density, a Genetic Programming (GP) system is proposed to evolve Bayesian-based programs. An evolved Bayesian-based program estimates the relevant multivariate density to construct a composite feature. The results of the experiments show that the GP system constructs high-level combined features which substantially improve the detection performance.

Genetic Programming Theory and Practice VI was developed from the sixth workshop at the University of Michigans Center for the Study of Complex Systems to facilitate the exchange of ideas and information related to the rapidly advancing field of Genetic Programming (GP). Contributions from the foremost international researchers and practitioners in the GP arena examine the similarities and differences between theoretical and empirical results on real-world problems. The text explores the synergy between theory and practice, producing a comprehensive view of the state of the art in GP application. These contributions address several significant interdependent themes which emerged from this years workshop, including: (1) Making efficient and effective use of test data. (2) Sustaining the long-term evolvability of our GP systems. (3) Exploiting discovered subsolutions for reuse. (4) Increasing the role of a Domain Expert.

The problem of identifying the trend in financial data series in order to forecast them for profit increase is addressed using genetic programming (GP). We enhance a GP system that searches for polynomial models of financial data series and relate it to a traditional GP manipulating functional models. Two of the key issues in the development are: 1) preprocessing of the series which includes data transformations and embedding; and 2) design of a proper fitness function that navigates the search by favouring parsimonious and predictive models. The two GP systems are applied for stock market analysis, and examined with real Tokyo Stock Exchange data. Using statistical and economical measures to estimate the results, we show that the GP could evolve profitable polynomials.

This report presents an asynchronous, distributed genetic programming (GP) system using a master/slave architecture. Using symbolic regression for fourier functions as the problem domain, the system was found to demonstrate cooperative coevolutionary dynamics when multiple client populations evolve solutions to similar, but different problems: specifically, closely coupled populations were found to promote continuous search, which in some cases leads to the discovery of better solutions.KeywordsGenetic AlgorithmCentral ServerGood OptimumRandom SolutionContinuous SearchThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

The author used J.R. Koza's (1992) genetic programming to evolve programs that classified contiguous regions of proteins as being /spl alpha/-helix cores or not. He snipped positive and negative examples of /spl alpha/-helix core regions out of a set of 90 proteins. These proteins were chosen from the Brookhaven Protein Data Bank to be non-homologous. The fitness of the programs was defined as the correlation coefficient between the observed and the predicted /spl alpha/-helicity of the above regions. The fittest program produced by the genetic programming system that predicted the training set at least as well as the testing set had a correlation of 0.4818 between the observed classifications and the classifications predicted by the program (on the proteins in the testing set). >

A blend of fuzzy logic and genetic programming is used in this research to achieve a single fine-tuned fuzzy rule, upon giving hundreds of fuzzy rules as the input. The system has Poisson arrival rate of vehicles, and decisions are taken to alter the sequence of lights based on the queue lengths of the lanes. The traffic simulator handles routing of vehicles in a single four-leg intersection with left and right turns. The fuzzy logic traffic controller system is used to generate the simulation data to feed the genetic programming system. The genetic programming system then creates an optimum fuzzy rule. This fine-tuned fuzzy rule is proven to be qualitatively better with respect to the mean square queue length and its error of the total system at any given point of time.