A aprendizagem incremental requer que o mecanismo de aprendizagem seja baseado no acúmulo dinâmico da informação extraída das experiências realizadas. A aprendizagem de máquina usando adaptatividade considera a integração de técnicas de aprendizagem de máquina simbólicas com técnicas adaptativas para a solução de problemas de aprendizagem. A palavra adaptatividade sugere a capacidade de modificação do conjunto de regras aprendidas em resposta a eventos que podem ocorrer durante o processo de aprendizagem, ou então autoajustes no conjunto de parâmetros. Os dispositivos adaptativos que possuem a capacidade de reter em suas regras informações extraídas de suas entradas podem acumular informações, para que sejam utilizadas quando forem necessárias. As estratégias de interesse para a incorporação da adaptatividade incluem a utilização de métodos e técnicas de aprendizagem de máquina, em particular as que implementam aprendizado supervisionado e tomada de decisão. O objetivo deste trabalho é explorar a utilização de técnicas adaptativas no processo de aprendizado por máquina, tanto de forma exclusiva como em conjunto com outras técnicas de aprendizagem. Para atingir este objetivo, propõe-se aqui a utilização de dispositivos adaptativos para representar o conhecimento adquirido através da aprendizagem incremental. Além disso, é feito um estudo de caso que combina aprendizagem de máquina com técnicas adaptativas para implementar um esquema de aprendizagem autônoma de estratégias, com o objetivo de vencer uma particular instância do jogo que é apresentado. A aprendizagem de um jogo exige a tomada de decisão, que é um processo complexo e dinâmico. Com a finalidade de fornecer um substrato geral para a criação, manipulação e análise de regras em problemas de tomada de decisão, utilizando tabelas de decisão adaptativas, a ferramenta de software Adapt-DT foi implementada. Um exemplo ilustrativo utilizando tabelas de decisão adaptativa como meio para a representação de conhecimento é apresentado, para exercitar a utilização da ferramenta. Isto permite concluir que os dispositivos adaptativos podem ser utilizados para representar o conhecimento adequadamente, com vantagens sobre outros métodos tradicionais. / Incremental learning requires a learning mechanism based on the information extracted from dynamically accumulated experiments. Adaptivity-oriented machine-learning combines adaptive techniques with symbolic ones for solving machine-learning problems. The term adaptivity means the ability of a learning process to change its own set of rules in response to events occurred during the learning process, or, equivalently, self-tuning the set of parameters. The adaptive devices with withhold information ability inside their rules, extracted from input from their own set of rules, can accumulate information to be used whenever they are necessary. The strategies of interest to adopt adaptivity include the use of machine learning techniques and methods, particularly the ones that implement supervised learning and decision-making. This work purposes to investigate the application of adaptive techniques in machine learning process, either exclusively and in cooperation with other techniques. In order to achieve this target, the use of adaptive devices to represent the knowledge gathered through incremental learning is proposed. Additionally, a case study that combines both machine learning and adaptive techniques to implement a scheme of autonomous learning strategies is also performed with the goal of winning an instance of the simple game. Decision-making is required to learning how to play a game, which is a complex and dynamic process. So as to provide a general framework for the creation, manipulation and analysis of rules in decision-making problems using adaptive decision tables, the Adapt-DT tool was implemented. An illustrative example using adaptive decision tables as a means to represent knowledge is introduced to the tool evaluation. This supports the conclusion that adaptive devices can be used to adequately represent the knowledge, with advantages over other traditional methods.

An algorithm of incremental mining implicative logical rules is pro-posed. This algorithm is based on constructing good classification tests. The in-cremental approach to constructing these rules allows revealing the interde-pendence between two fundamental components of human thinking: pattern recognition and knowledge acquisition.

Cette thèse s’intéresse à la problématique de la classification dynamique en environnements stationnaires et non stationnaires, tolérante aux variations de quantités des données d’apprentissage et capable d’ajuster ses modèles selon la variabilité des données entrantes. Pour cela, nous proposons une solution faisant cohabiter des classificateurs one-class SVM indépendants ayant chacun leur propre procédure d’apprentissage incrémentale et par conséquent, ne subissant pas d’influences croisées pouvant émaner de la configuration des modèles des autres classificateurs. L’originalité de notre proposition repose sur l’exploitation des anciennes connaissances conservées dans les modèles de SVM (historique propre à chaque SVM représenté par l’ensemble des vecteurs supports trouvés) et leur combinaison avec les connaissances apportées par les nouvelles données au moment de leur arrivée. Le modèle de classification proposé (mOC-iSVM) sera exploité à travers trois variations exploitant chacune différemment l’historique des modèles. Notre contribution s’inscrit dans un état de l’art ne proposant pas à ce jour de solutions permettant de traiter à la fois la dérive de concepts, l’ajout ou la suppression de concepts, la fusion ou division de concepts, tout en offrant un cadre privilégié d’interactions avec l’utilisateur. Dans le cadre du projet ANR DIGIDOC, notre approche a été appliquée sur plusieurs scénarios de classification de flux d’images pouvant survenir dans des cas réels lors de campagnes de numérisation. Ces scénarios ont permis de valider une exploitation interactive de notre solution de classification incrémentale pour classifier des images arrivant en flux afin d’améliorer la qualité des images numérisées. / This research contributes to the field of dynamic learning and classification in case of stationary and non-stationary environments. The goal of this PhD is to define a new classification framework able to deal with very small learning dataset at the beginning of the process and with abilities to adjust itself according to the variability of the incoming data inside a stream. For that purpose, we propose a solution based on a combination of independent one-class SVM classifiers having each one their own incremental learning procedure. Consequently, each classifier is not sensitive to crossed influences which can emanate from the configuration of the models of the other classifiers. The originality of our proposal comes from the use of the former knowledge kept in the SVM models (represented by all the found support vectors) and its combination with the new data coming incrementally from the stream. The proposed classification model (mOC-iSVM) is exploited through three variations in the way of using the existing models at each step of time. Our contribution states in a state of the art where no solution is proposed today to handle at the same time, the concept drift, the addition or the deletion of concepts, the fusion or division of concepts while offering a privileged solution for interaction with the user. Inside the DIGIDOC project, our approach was applied to several scenarios of classification of images streams which can correspond to real cases in digitalization projects. These different scenarios allow validating an interactive exploitation of our solution of incremental classification to classify images coming in a stream in order to improve the quality of the digitized images.

An algorithm of incremental mining implicative logical rules is pro-posed. This algorithm is based on constructing good classification tests. The in-cremental approach to constructing these rules allows revealing the interde-pendence between two fundamental components of human thinking: pattern recognition and knowledge acquisition.

Designing object-oriented software is hard, and designing reusable object-oriented software is even harder. This task is even more daunting for a developer of computational intelligence applications, as optimising one design objective tends to make others inefficient or even impossible. Classic examples in computer science include ‘storage vs. time’ and ‘simplicity vs. flexibility.’ Neural network requirements are by their very nature very tightly coupled – a required design change in one area of an existing application tends to have severe effects in other areas, making the change impossible or inefficient. Often this situation leads to a major redesign of the system and in many cases a completely rewritten application. Many commercial and open-source packages do exist, but these cannot always be extended to support input from other fields of computational intelligence due to proprietary reasons or failing to fully take all design requirements into consideration. Design patterns make a science out of writing software that is modular, extensible and efficient as well as easy to read and understand. The essence of a design pattern is to avoid repeatedly solving the same design problem from scratch by reusing a solution that solves the core problem. This pattern or template for the solution has well understood prerequisites, structure, properties, behaviour and consequences. CILib is a framework that allows developers to develop new computational intelligence applications quickly and efficiently. Flexibility, reusability and clear separation between components are maximised through the use of design patterns. Reliability is also ensured as the framework is open source and thus has many people that collaborate to ensure that the framework is well designed and error free. This dissertation discusses the design and implementation of a generic neural network framework that allows users to design, implement and use any possible neural network models and algorithms in such a way that they can reuse and be reused by any other computational intelligence algorithm in the rest of the framework, or any external applications. This is achieved by using object-oriented design patterns in the design of the framework. / Dissertation (MSc)--University of Pretoria, 2007. / Computer Science / unrestricted

Medical doctors understaffing is becoming a compelling problem in many healthcare systems. This problem can be alleviated by utilising Computer-Aided Diagnosis (CAD) systems to substitute doctors in different tasks, for instance, histopa-thological image classification. The recent surge of deep learning has allowed CAD systems to perform this task at a very competitive performance. However, a major challenge with this task is the need to periodically update the models with new data and/or new classes or diseases. These periodical updates will result in catastrophic forgetting, as Convolutional Neural Networks typically requires the entire data set beforehand and tend to lose knowledge about old data when trained on new data. Incremental learning methods were proposed to alleviate this problem with deep learning. In this thesis, two incremental learning methods, Learning without Forgetting (LwF) and a generative rehearsal-based method, are investigated. They are evaluated on two criteria: The first, capability of incrementally adding new classes to a pre-trained model, and the second is the ability to update the current model with an new unbalanced data set. Experiments shows that LwF does not retain knowledge properly for the two cases. Further experiments are needed to draw any definite conclusions, for instance using another training approach for the classes and try different combinations of losses. On the other hand, the generative rehearsal-based method tends to work for one class, showing a good potential to work if better quality images were generated. Additional experiments are also required in order to investigating new architectures and approaches for a more stable training.

Feature extraction is an extremely important pre-processing step to pattern recognition, and machine learning problems. This thesis highlights how one can best extract features from the data in an exhaustively online and purely adaptive manner. The solution to this problem is given for both labeled and unlabeled datasets, by presenting a number of novel on-line learning approaches. Specifically, the differential equation method for solving the generalized eigenvalue problem is used to derive a number of novel machine learning and feature extraction algorithms. The incremental eigen-solution method is used to derive a novel incremental extension of linear discriminant analysis (LDA). Further the proposed incremental version is combined with extreme learning machine (ELM) in which the ELM is used as a preprocessor before learning. In this first key contribution, the dynamic random expansion characteristic of ELM is combined with the proposed incremental LDA technique, and shown to offer a significant improvement in maximizing the discrimination between points in two different classes, while minimizing the distance within each class, in comparison with other standard state-of-the-art incremental and batch techniques. In the second contribution, the differential equation method for solving the generalized eigenvalue problem is used to derive a novel state-of-the-art purely incremental version of slow feature analysis (SLA) algorithm, termed the generalized eigenvalue based slow feature analysis (GENEIGSFA) technique. Further the time series expansion of echo state network (ESN) and radial basis functions (EBF) are used as a pre-processor before learning. In addition, the higher order derivatives are used as a smoothing constraint in the output signal. Finally, an online extension of the generalized eigenvalue problem, derived from James Stone’s criterion, is tested, evaluated and compared with the standard batch version of the slow feature analysis technique, to demonstrate its comparative effectiveness. In the third contribution, light-weight extensions of the statistical technique known as canonical correlation analysis (CCA) for both twinned and multiple data streams, are derived by using the same existing method of solving the generalized eigenvalue problem. Further the proposed method is enhanced by maximizing the covariance between data streams while simultaneously maximizing the rate of change of variances within each data stream. A recurrent set of connections used by ESN are used as a pre-processor between the inputs and the canonical projections in order to capture shared temporal information in two or more data streams. A solution to the problem of identifying a low dimensional manifold on a high dimensional dataspace is then presented in an incremental and adaptive manner. Finally, an online locally optimized extension of Laplacian Eigenmaps is derived termed the generalized incremental laplacian eigenmaps technique (GENILE). Apart from exploiting the benefit of the incremental nature of the proposed manifold based dimensionality reduction technique, most of the time the projections produced by this method are shown to produce a better classification accuracy in comparison with standard batch versions of these techniques - on both artificial and real datasets.

Nous présentons deux contributions majeures : 1) une combinaison de plusieurs descripteurs d’images pour la classification à grande échelle, 2) des algorithmes parallèles de SVM pour la classification d’images à grande échelle. Nous proposons aussi un algorithme incrémental et parallèle de classification lorsque les données ne peuvent plus tenir en mémoire vive. / We have proposed a novel method of combination multiple of different features for image classification. For large scale learning classifiers, we have developed the parallel versions of both state-of-the-art linear and nonlinear SVMs. We have also proposed a novel algorithm to extend stochastic gradient descent SVM for large scale learning. A class of large scale incremental SVM classifiers has been developed in order to perform classification tasks on large datasets with very large number of classes and training data can not fit into memory.

Métodos baseados em grafos consistem em uma poderosa forma de representação e abstração de dados que proporcionam, dentre outras vantagens, representar relações topológicas, visualizar estruturas, representar grupos de dados com formatos distintos, bem como, fornecer medidas alternativas para caracterizar os dados. Esse tipo de abordagem tem sido cada vez mais considerada para solucionar problemas de aprendizado de máquina, principalmente no aprendizado não supervisionado, como agrupamento de dados, e mais recentemente, no aprendizado semissupervisionado. No aprendizado supervisionado, por outro lado, o uso de algoritmos baseados em grafos ainda tem sido pouco explorado na literatura. Este trabalho apresenta um algoritmo não paramétrico baseado em grafos para problemas de classificação com distribuição estacionária, bem como sua extensão para problemas que apresentam distribuição não estacionária. O algoritmo desenvolvido baseia-se em dois conceitos, a saber, 1) em uma estrutura chamada grafo K-associado ótimo, que representa o conjunto de treinamento como um grafo esparso e dividido em componentes; e 2) na medida de pureza de cada componente, que utiliza a estrutura do grafo para determinar o nível de mistura local dos dados em relação às suas classes. O trabalho também considera problemas de classificação que apresentam alteração na distribuição de novos dados. Este problema caracteriza a mudança de conceito e degrada o desempenho do classificador. De modo que, para manter bom desempenho, é necessário que o classificador continue aprendendo durante a fase de aplicação, por exemplo, por meio de aprendizado incremental. Resultados experimentais sugerem que ambas as abordagens apresentam vantagens na classificação de dados em relação aos algoritmos testados / Graph-based methods consist in a powerful form for data representation and abstraction which provides, among others advantages, representing topological relations, visualizing structures, representing groups of data with distinct formats, as well as, supplying alternative measures to characterize data. Such approach has been each time more considered to solve machine learning related problems, mainly concerning unsupervised learning, like clustering, and recently, semi-supervised learning. However, graph-based solutions for supervised learning tasks still remain underexplored in literature. This work presents a non-parametric graph-based algorithm suitable for classification problems with stationary distribution, as well as its extension to cope with problems of non-stationary distributed data. The developed algorithm relies on the following concepts, 1) a graph structure called optimal K-associated graph, which represents the training set as a sparse graph separated into components; and 2) the purity measure for each component, which uses the graph structure to determine local data mixture level in relation to their classes. This work also considers classification problems that exhibit modification on distribution of data flow. This problem qualifies concept drift and worsens any static classifier performance. Hence, in order to maintain accuracy performance, it is necessary for the classifier to keep learning during application phase, for example, by implementing incremental learning. Experimental results, concerning both algorithms, suggest that they had presented advantages over the tested algorithms on data classification tasks

Este trabalho propõe uma nova abordagem conexionista, chamada de IGMN (do inglês Incremental Gaussian Mixture Network), para aproximação incremental de funções e tarefas de tempo real. Ela é inspirada em recentes teorias do cérebro, especialmente o MPF (do inglês Memory-Prediction Framework) e a Inteligência Artificial Construtivista, que fazem com que o modelo proposto possua características especiais que não estão presentes na maioria dos modelos de redes neurais existentes. Além disso, IGMN é baseado em sólidos princípios estatísticos (modelos de mistura gaussianos) e assintoticamente converge para a superfície de regressão ótima a medida que os dados de treinamento chegam. As principais vantagens do IGMN em relação a outros modelos de redes neurais são: (i) IGMN aprende instantaneamente analisando cada padrão de treinamento apenas uma vez (cada dado pode ser imediatamente utilizado e descartado); (ii) o modelo proposto produz estimativas razoáveis baseado em poucos dados de treinamento; (iii) IGMN aprende de forma contínua e perpétua a medida que novos dados de treinamento chegam (não existem fases separadas de treinamento e utilização); (iv) o modelo proposto resolve o dilema da estabilidade-plasticidade e não sofre de interferência catastrófica; (v) a topologia da rede neural é definida automaticamente e de forma incremental (novas unidades são adicionadas sempre que necessário); (vi) IGMN não é sensível às condições de inicialização (de fato IGMN não utiliza nenhuma decisão e/ou inicialização aleatória); (vii) a mesma rede neural IGMN pode ser utilizada em problemas diretos e inversos (o fluxo de informações é bidirecional) mesmo em regiões onde a função alvo tem múltiplas soluções; e (viii) IGMN fornece o nível de confiança de suas estimativas. Outra contribuição relevante desta tese é o uso do IGMN em importantes tarefas nas áreas de robótica e aprendizado de máquina, como por exemplo a identificação de modelos, a formação incremental de conceitos, o aprendizado por reforço, o mapeamento robótico e previsão de séries temporais. De fato, o poder de representação e a eficiência e do modelo proposto permitem expandir o conjunto de tarefas nas quais as redes neurais podem ser utilizadas, abrindo assim novas direções nos quais importantes contribuições do estado da arte podem ser feitas. Através de diversos experimentos, realizados utilizando o modelo proposto, é demonstrado que o IGMN é bastante robusto ao problema de overfitting, não requer um ajuste fino dos parâmetros de configuração e possui uma boa performance computacional que permite o seu uso em aplicações de controle em tempo real. Portanto pode-se afirmar que o IGMN é uma ferramenta de aprendizado de máquina bastante útil em tarefas de aprendizado incremental de funções e predição em tempo real. / This work proposes IGMN (standing for Incremental Gaussian Mixture Network), a new connectionist approach for incremental function approximation and real time tasks. It is inspired on recent theories about the brain, specially the Memory-Prediction Framework and the Constructivist Artificial Intelligence, which endows it with some unique features that are not present in most ANN models such as MLP, RBF and GRNN. Moreover, IGMN is based on strong statistical principles (Gaussian mixture models) and asymptotically converges to the optimal regression surface as more training data arrive. The main advantages of IGMN over other ANN models are: (i) IGMN learns incrementally using a single scan over the training data (each training pattern can be immediately used and discarded); (ii) it can produce reasonable estimates based on few training data; (iii) the learning process can proceed perpetually as new training data arrive (there is no separate phases for leaning and recalling); (iv) IGMN can handle the stability-plasticity dilemma and does not suffer from catastrophic interference; (v) the neural network topology is defined automatically and incrementally (new units added whenever is necessary); (vi) IGMN is not sensible to initialization conditions (in fact there is no random initialization/ decision in IGMN); (vii) the same neural network can be used to solve both forward and inverse problems (the information flow is bidirectional) even in regions where the target data are multi-valued; and (viii) IGMN can provide the confidence levels of its estimates. Another relevant contribution of this thesis is the use of IGMN in some important state-of-the-art machine learning and robotic tasks such as model identification, incremental concept formation, reinforcement learning, robotic mapping and time series prediction. In fact, the efficiency of IGMN and its representational power expand the set of potential tasks in which the neural networks can be applied, thus opening new research directions in which important contributions can be made. Through several experiments using the proposed model it is demonstrated that IGMN is also robust to overfitting, does not require fine-tunning of its configuration parameters and has a very good computational performance, thus allowing its use in real time control applications. Therefore, IGMN is a very useful machine learning tool for incremental function approximation and on-line prediction.