Optimization in an Error Backpropagation Neural Network Environment with a Performance Test on a Spectral Pattern Classification Problem

This paper attempts to develop a mathematically rigid framework for minimizing the cross-entropy function in an error backpropagating framework. In doing so, we derive the backpropagation formulae for evaluating the partial derivatives in a computationally efficient way. Various techniques of optimizing the multiple-class cross-entropy error function to train single hidden layer neural network classifiers with softmax output transfer functions are investigated on a real-world multispectral pixel-by-pixel classification problem that is of fundamental importance in remote sensing. These techniques include epoch-based and batch versions of backpropagation of gradient descent, PR-conjugate gradient, and BPGS quasi-Newton errors. The method of choice depends upon the nature of the learning task and whether one wants to optimize learning for speed or classification performance. It was found that, comparatively considered, gradient descent error backpropagation provided the best and most stable out-of-sample performance results across batch and epoch-based modes of operation. If the goal is to maximize learning speed and a sacrifice in classification accuracy is acceptable, then PR-conjugate gradient error backpropagation tends to be superior. If the training set is very large, stochastic epoch-based versions of local optimizers should be chosen utilizing a larger rather than a smaller epoch size to avoid unacceptable instabilities in the classification results.     

Assessing Activity Pattern Similarity with Multidimensional Sequence Alignment Based on a Multiobjective Optimization Evolutionary Algorithm

Due to the complexity and multidimensional characteristics of human activities, assessing the similarity of human activity patterns and classifying individuals with similar patterns remains highly challenging. This article presents a new and unique methodology for evaluating the similarity among individual activity patterns. It conceptualizes multidimensional sequence alignment as a multiobjective optimization problem and solves this problem with an evolutionary algorithm (EA). The study utilizes sequence alignment to code multiple facets of human activities into multidimensional sequences and to treat similarity assessment as a multiobjective optimization problem that aims to minimize the alignment cost for all dimensions simultaneously. A multiobjective optimization evolutionary algorithm is used to generate a diverse set of optimal or near‐optimal alignment solutions. Evolutionary operators are specifically designed for this problem, and a local search method is also incorporated to improve the search ability of the algorithm. We demonstrate the effectiveness of our method by comparing it with a popular existing method called ClustalG using a set of 50 sequences. The results indicate that our method outperforms the existing method for most of our selected cases. The multiobjective EA presented in this article provides an effective approach for assessing activity pattern similarity and a foundation for identifying distinctive groups of individuals with similar activity patterns. Debido a la complejidad y el carácter multidimensional de las actividades humanas, la evaluación de la similitud de los patrones de la actividad humana y la clasificación de individuos con patrones similares sigue siendo una tarea muy difícil. Este artículo presenta una metodología novedosa y única para evaluar la similitud entre los patrones de actividad individual. Se conceptualiza el método de alineación de secuencias multidimensionales (Multidimensional Sequence Alignment‐MDSA) como un problema de optimización multiobjetivo y se resuelve mediante un algoritmo evolutivo (evolutionary algorithm‐EA). El estudio utiliza la alineación de secuencias para codificar múltiples facetas de las actividades humanas en secuencias multidimensionales y para tratar la evaluación de similitud como un problema de optimización multiobjetivo que tiene como objetivo reducir al mínimo el coste de alineación para todas las dimensiones simultáneamente . Los autores utilizan algoritmo evolutivo de optimización multiobjetivo (multiobjective optimization evolutionary algorithm‐MOEA) para generar un conjunto diverso de soluciones de alineación óptimas o casi óptimas. Los operadores evolutivos del algoritmo están diseñados específicamente para este problema, al cual se le incorpora un método de búsqueda local para mejorar su desempeño. Los autores demuestran la eficacia del método propuesto mediante su comparación al método existente ClustalG haciendo uso de un conjunto de 50 secuencias. Los resultados indican que el método propuesto supera a los métodos existentes para la mayoría de casos seleccionados. La EA multiobjetivo presentado en este artículo proporciona un método eficaz para evaluar similitudes del patrón de actividad y constituye un componente fundamental para la identificación de grupos distintivos de personas con patrones de actividad similares. 由于人类活动的复杂性和多维度特性,人类活动模式的相似性评估和基于相似模式的个体分类仍具有较高的挑战性。本文提出了一种新型的、独特的方法来评估个体活动模式的相似性。该研究利用序列比对将人类活动的多个方面编码成多维序列,并将活动模式相似性评估作为一个多目标优化问题,其目的是使所有维度的比对成本同时最低。多目标最优进化算法被用于生成一组不同的最优或接近最优的比对解决方案。为此特别设计了进化算子,并集成一种局部搜索方法提高了算法的整体搜索能力。为验证该方法的有效性,本文以50对序列作为实验,并与常用的ClustalG方法进行了对比。研究结果表明,该方法在多数选取案例中性能优于ClustalG方法。因此,本文提出的多目标进化（EA）算法为个体活动模式相似性评估提供了一种有效方法,也为具有相似活动模式个体中特定群体识别奠定了基础。     

Experiencing Waterholes in an Arid Environment,with Particular Reference to the Lake Eyre Basin,Australia: a Review

The distribution and permanence of waterholes have dictated human activity and settlement patterns in the arid zone for millennia. At a fundamental level, waterholes are vital to survival in a harsh climate. The recreational value and use of waterholes are widely recognised; however, the less tangible aspects of ‘waterhole experience’ have received scant attention. This review, written whilst compiling an inventory of permanent waterbodies across the Lake Eyre Basin of Queensland and South Australia, aimed to provide a deeper understanding of the interaction of humans with water in an arid land, primarily from a settler perspective. A large body of literature pertaining to waterholes was reviewed, including Aboriginal histories, explorer journals, accounts of early settlers, and numerous fictional and factual accounts of life in inland Australia. It becomes apparent that human appreciation of arid-zone waterholes runs deeper than the survival imperative, and can be enriched by considering landscape aesthetics and psychology. This attraction has spawned a rich mythology centred on waterholes – permanent features in an otherwise harsh and variable environment.