Marcus Gallagher

Experimental results for the special session on real-parameter optimization at CEC 2005: a simple, continuous EDA

A comprehensive set of experiments was conducted with a continuous EDA on 25 test problems provided in the real-parameter optimization special session. It is expected that the results presented here could be used to gain some deeper understanding of the performance of the EDA as well as facilitate the comparison across different algorithms.

On the importance of diversity maintenance in estimation of distribution algorithms

The development of Estimation of Distribution Algorithms (EDAs) has largely been driven by using more and more complex statistical models to approximate the structure of search space. However, there are still problems that are difficult for EDAs even with models capable of capturing high order dependences. In this paper, we show that diversity maintenance plays an important role in the performance of EDAs. A continuous EDA based on the Cholesky decomposition is tested on some well-known difficult benchmark problems to demonstrate how different diversity maintenance approaches could be applied to substantially improve its performance.

A general-purpose tunable landscape generator

The research literature on metaheuristic and evolutionary computation has proposed a large number of algorithms for the solution of challenging real-world optimization problems. It is often not possible to study theoretically the performance of these algorithms unless significant assumptions are made on either the algorithm itself or the problems to which it is applied, or both. As a consequence, metaheuristics are typically evaluated empirically using a set of test problems. Unfortunately, relatively little attention has been given to the development of methodologies and tools for the large-scale empirical evaluation and/or comparison of metaheuristics. In this paper, we propose a landscape (test-problem) generator that can be used to generate optimization problem instances for continuous, bound-constrained optimization problems. The landscape generator is parameterized by a small number of parameters, and the values of these parameters have a direct and intuitive interpretation in terms of the geometric features of the landscapes that they produce. An experimental space is defined over algorithms and problems, via a tuple of parameters for any specified algorithm and problem class (here determined by the landscape generator). An experiment is then clearly specified as a point in this space, in a way that is analogous to other areas of experimental algorithmics, and more generally in experimental design. Experimental results are presented, demonstrating the use of the landscape generator. In particular, we analyze some simple, continuous estimation of distribution algorithms, and gain new insights into the behavior of these algorithms using the landscape generator

Statistical Racing Techniques for Improved Empirical Evaluation of Evolutionary Algorithms

In empirical studies of Evolutionary Algorithms, it is usually desirable to evaluate and compare algorithms using as many different parameter settings and test problems as possible, in order to have a clear and detailed picture of their performance. Unfortunately, the total number of experiments required may be very large, which often makes such research work computationally prohibitive. In this paper, the application of a statistical method called racing is proposed as a general-purpose tool to reduce the computational requirements of large-scale experimental studies in evolutionary algorithms. Experimental results are presented that show that racing typically requires only a small fraction of the cost of an exhaustive experimental study.

Intelligent Data Engineering and Automated Learning - IDEAL 2005

Data Mining and Knowledge Engineering.- EXiT-B: A New Approach for Extracting Maximal Frequent Subtrees from XML Data.- Synthetic Environment Representational Semantics Using the Web Ontology Language.- New Rules for Hybrid Spatial Reasoning.- Using Pre-aggregation for Efficient Spatial Query Processing in Sensor Environments.- Model Trees for Classification of Hybrid Data Types.- Finding Uninformative Features in Binary Data.- Knowledge Reduction of Rough Set Based on Partition.- Multiresolution Analysis of Connectivity.- Kernel Biased Discriminant Analysis Using Histogram Intersection Kernel for Content-Based Image Retrieval.- Unsupervised Image Segmentation Using Penalized Fuzzy Clustering Algorithm.- Multi-attributes Image Analysis for the Classification of Web Documents Using Unsupervised Technique.- Automatic Image Annotation Based on Topic-Based Smoothing.- A Focused Crawler with Document Segmentation.- An Intelligent Grading System Using Heterogeneous Linguistic Resources.- Probabilistic Data Generation for Deduplication and Data Linkage.- Mining Job Logs Using Incremental Attribute-Oriented Approach.- Dimensional Reduction of Large Image Datasets Using Non-linear Principal Components.- Classification by Instance-Based Learning Algorithm.- Analysis/Synthesis of Speech Signals Based on AbS/OLA Sinusoidal Modeling Using Elliptic Filter.- Robust Model Adaptation Using Mean and Variance Transformations in Linear Spectral Domain.- Using Support Vector Machine for Modeling of Pulsed GTAW Process.- Design of Simple Structure Neural Voltage Regulator for Power Systems.- EEG Source Localization for Two Dipoles in the Brain Using a Combined Method.- Intelligent Control of Micro Heat Exchanger with Locally Linear Identifier and Emotional Based Controller.- Identification of Anomalous SNMP Situations Using a Cooperative Connectionist Exploratory Projection Pursuit Model.- Learning Algorithms and Systems.- Neural Networks: A Replacement for Gaussian Processes?.- A Dynamic Merge-or-Split Learning Algorithm on Gaussian Mixture for Automated Model Selection.- Bayesian Radial Basis Function Neural Network.- An Empirical Study of Hoeffding Racing for Model Selection in k-Nearest Neighbor Classification.- Designing an Optimal Network Using the Cross-Entropy Method.- Generating Predicate Rules from Neural Networks.- Improving Ensembles with Classificational Cellular Automata.- A Gradient BYY Harmony Learning Algorithm on Mixture of Experts for Curve Detection.- A Novel Anomaly Detection Using Small Training Sets.- Induction of Linear Decision Trees with Real-Coded Genetic Algorithms and k-D Trees.- Intelligent Predictive Control of a 6-Dof Robotic Manipulator with Reliability Based Performance Improvement.- Sequential Search for Decremental Edition.- Bearing Similarity Measures for Self-organizing Feature Maps.- Efficient Spatial Clustering Algorithm Using Binary Tree.- Cluster Analysis of High-Dimensional Data: A Case Study.- Universal Clustering with Family of Power Loss Functions in Probabilistic Space.- Circular SOM for Temporal Characterisation of Modelled Gene Expressions.- Recursive Self-organizing Map as a Contractive Iterative Function System.- Differential Priors for Elastic Nets.- Graphics Hardware Implementation of the Parameter-Less Self-organising Map.- Weighted SOM-Face: Selecting Local Features for Recognition from Individual Face Image.- SOM-Based Novelty Detection Using Novel Data.- Multi-level Document Classifications with Self-organising Maps.- Bioinformatics.- Predictive Vaccinology: Optimisation of Predictions Using Support Vector Machine Classifiers.- Evolving Neural Networks for the Classification of Malignancy Associated Changes.- Matching Peptide Sequences with Mass Spectra.- Extraction by Example: Induction of Structural Rules for the Analysis of Molecular Sequence Data from Heterogeneous Sources.- A Multi-population ? 2 Test Approach to Informative Gene Selection.- Gene Selection of DNA Microarray Data Based on Regularization Networks.- Application of Mixture Models to Detect Differentially Expressed Genes.- A Comparative Study of Two Novel Predictor Set Scoring Methods.- Deriving Matrix of Peptide-MHC Interactions in Diabetic Mouse by Genetic Algorithm.- SVM Based Prediction of Bacterial Transcription Start Sites.- Exploiting Sequence Dependencies in the Prediction of Peroxisomal Proteins.- Protein Fold Recognition Using Neural Networks and Support Vector Machines.- Agents and Complex Systems.- Support Tool for Multi-agent Development.- A Hybrid Agent Architecture for Modeling Autonomous Agents in SAGE.- Toward Transitive Dependence in MAS.- An Architecture for Multi-agent Based Self-adaptive System in Mobile Environment.- Autonomous and Dependable Recovery Scheme in UPnP Network Settings.- A Transitive Dependence Based Social Reasoning Mechanism for Coalition Formation.- A Multi-agent Based Context Aware Self-healing System.- Combining Influence Maps and Cellular Automata for Reactive Game Agents.- Patterns in Complex Systems Modeling.- Global Optimization Using Evolutionary Algorithm Based on Level Set Evolution and Latin Square.- Co-evolutionary Rule-Chaining Genetic Programming.- A Dynamic Migration Model for Self-adaptive Genetic Algorithms.- Financial Engineering.- A Multicriteria Sorting Procedure for Financial Classification Problems: The Case of Business Failure Risk Assessment.- Volatility Modelling of Multivariate Financial Time Series by Using ICA-GARCH Models.- Volatility Transmission Between Stock and Bond Markets: Evidence from US and Australia.- A Machine Learning Approach to Intraday Trading on Foreign Exchange Markets.

Evolving Pac-Man Players: Can We Learn from Raw Input?

Pac-Man (and variant) computer games have received some recent attention in artificial intelligence research. One reason is that the game provides a platform that is both simple enough to conduct experimental research and complex enough to require non-trivial strategies for successful game-play. This paper describes an approach to developing Pac-Man playing agents that learn game-play based on minimal onscreen information. The agents are based on evolving neural network controllers using a simple evolutionary algorithm. The results show that neuroevolution is able to produce agents that display novice playing ability, with a minimal amount of onscreen information, no knowledge of the rules of the game and a minimally informative fitness function. The limitations of the approach are also discussed, together with possible directions for extending the work towards producing better Pac-Man playing agents

Playing in continuous spaces: some analysis and extension of population-based incremental learning

As an alternative to traditional evolutionary algorithms (EAs), population-based incremental learning (PBIL) maintains a probabilistic model of the best individual(s). Originally, PBIL was applied in binary search spaces. Recently, some work has been done to extend it to continuous spaces. In this paper, we review two such extensions of PBIL. An improved version of the PBIL based on Gaussian model is proposed that combines two main features: a new updating rule that takes into account all the individuals and their fitness values and a self-adaptive learning rate parameter. Furthermore, a new continuous PBIL employing a histogram probabilistic model is proposed. Some experiments results are presented that highlight the features of the new algorithms.

An influence map model for playing Ms. Pac-Man

In this paper we develop a Ms. Pac-Man playing agent based on an influence map model. The proposed model is as simple as possible while capturing the essentials of the game. Our model has three main parameters that have an intuitive relationship to the agents behavior. Experimental results are presented exploring the models performance over its parameter space using random and systematic global exploration and a greedy algorithm. The model parameters can be optimized without difficulty despite the noisy fitness function used. The performance of the optimized agents is comparable to the best published results for a Ms. Pac-Man playing agent. Nevertheless, some difficulties were observed in terms of the model and the software system.

Population-Based Continuous Optimization, Probabilistic Modelling and Mean Shift

Evolutionary algorithms perform optimization using a population of sample solution points. An interesting development has been to view population-based optimization as the process of evolving an explicit, probabilistic model of the search space. This paper investigates a formal basis for continuous, population-based optimization in terms of a stochastic gradient descent on the Kullback-Leibler divergence between the model probability density and the objective function, represented as an unknown density of assumed form. This leads to an update rule that is related and compared with previous theoretical work, a continuous version of the population-based incremental learning algorithm, and the generalized mean shift clustering framework. Experimental results are presented that demonstrate the dynamics of the new algorithm on a set of simple test problems.

Combining Meta-EAs and racing for difficult EA parameter tuning tasks

This chapter presents a novel framework for tuning the parameters of Evolutionary Algorithms. A hybrid technique combining Meta-EAs and statistical Racing approaches is developed, which is not only capable of effectively exploring the search space of numerical parameters but also suitable for tuning symbolic parameters where it is generally difficult to define any sensible distance metric.