This master's thesis offers a brief introduction to evolutionary computation. It describes and compares the genetic programming and grammar based genetic programming and their potential use in automatic software repair. It studies possible applications of grammar based genetic programming on automatic software repair. Grammar based genetic programming is then used in design and implementation of a new method for automatic software repair. Experimental evaluation of the implemented automatic repair was performed on set of test programs.

[pt] A programação de produção em refinaria pode ser compreendida como decisões que buscam otimizar alocação de recursos, o sequenciamento de atividades e a sua realização temporal, respeitando restrições e visando ao atendimento de múltiplos objetivos. Apesar da complexidade e natureza combinatória, a atividade carece de sistemas sofisticados que auxiliem o processo decisório, especialmente baseadas em otimização, pois as ferramentas utilizadas são planilhas ou softwares de simulação. A diversidade de objetivos do problema não implica em equivalência de importância. Pode-se considerar que existem grupos, onde os que afetam diretamente a capacidade produtiva da refinaria se sobrepõem aos associados à maior continuidade operacional. Esta tese propõe o desenvolvimento de algoritmos multiobjetivos para programação de petróleo em refinaria. As propostas se baseiam em conceituadas técnicas da literatura multiobjetivo, como dominância de Pareto e decomposição do problema, integradas à programação genética com inspiração quântica. São estudados modelos em um ou dois níveis de decisão. A diferenciação dos grupos de objetivos é avaliada com base em critérios estabelecidos para considerar uma solução proposta como aceitável e também é avaliada a influência de uma população externa no processo evolutivo. Os modelos são testados em cenários de uma refinaria real e os resultados são comparados com um modelo que trata os objetivos de forma hierarquizada. As abordagens baseadas em dominância e em decomposição apresentam vantagem sobre o algoritmo hierarquizado, e a decomposição é superior. Numa comparação com o modelo em dois níveis de decisão, apenas o que utiliza estratégia de decomposição em cada nível apresenta bons resultados. Ao final deste trabalho é obtido mais de um modelo multiobjetivo capaz de oferecer um conjunto de soluções que atendam aos objetivos críticos e deem flexibilidade de análise a posteriori para o programador de produção, o que, por exemplo, permite que ele pondere questões não mapeadas no modelo. / [en] Refinery scheduling can be understood as a set of decisions which aims to optimize resource allocation, task sequencing, and their time-related execution, respecting constraints and targeting multiple objectives. Despite its complexity and combinatorial nature, the refinery scheduling lacks more sophisticated support decision tools. The main systems in the area are worksheets and, sometimes, simulation software. The multiple objectives do not mean they have the same importance. Actually, they can be grouped whereas the objectives related to the refinery production capacity are more important than the ones related to a smooth operation. This thesis proposes the development of multiobjective algorithms applied to crude oil refinery scheduling. The proposals are based on the major technics of multiobjective literature, like Pareto dominance and problem decomposition, integrated with a quantum-inspired genetic programming approach. One and two decision level models are studied. The difference between groups is handled with conditions that define what can be considered a good solution. The effect of using an archive population in the evolutionary process is also evaluated. The results of the proposed models are compared with another model that handles the objectives in a hierarchical logical. Both decomposition and dominance approaches have better results than the hierarchical model. The decomposition model is even better. The bilevel decomposition method is the only one, among two decision levels models, which have shown good performance. In the end, this work achieves more than one multiobjective model able to offer a set of solutions which comprises the critical objectives and can give flexibility to the production scheduler does his analysis. Therefore, he can consider aspects not included in the model, like the forecast of crude oil batches not scheduled yet.

在目前的企業環境中，很多企業致力於管理和應用組織知識，來維持他們的核心能力和創造競爭優勢。有效率的管理組織知識，能減少解決問題的時間和成本，並增加組織學習和創新能力。並且，由於累積知識資源的需求，很多企業開始發展知識庫，以儲存組織及個人的知識，用來增加組織整體的效率、支援日常的運作以及企業策略的操作。 知識管理是現代的典範，可用來有效管理組織知識，進而改善組織績效。知識管理的目的是強調管理知識的流動及流程。在知識管理流程方面，主要區分為知識擷取、整合、儲存/歸類、散播和應用知識等程序。另外，資訊技術可用來協助知識管理，並能使知識管理更有效率。知識管理的主要議題之ㄧ是知識的擷取，由於目前知識來源的提供，主要是透過知識工作者，可是它對於知識工作者而言，是一種額外的負擔。因此，設計一個有效的方法來自動產生組織知識，以減輕他們的額外負擔，將是一個很重要的課題。第二個相當重要的議題是知識整合，由於不同來源的知識，可能造成組織知識的衝突，因此設計一個知識整合方法，把不同來源的知識整合成一個完整的知識，組織將會逐漸增加這方面的需求。 分類在很多應用中是常遭遇的問題，例如財務預測、疾病診斷等。在過去，分類規則常藉由決策樹的方法所產生，並用於解決分類的問題。在本論文中，提出兩個以遺傳規劃為基礎的知識擷取方法和兩個以遺傳規劃為基礎的知識整合方法，分別支援知識管理流程中的知識擷取和知識整合。 在兩個所提的知識擷取方法中，第一個方法是著重在快速和容易地找到想要的分類樹，但是，此方法可能會產生結構較複雜的分類樹。第二個方法是修正第一個方法，產生一個較精簡和應用性高的分類樹。這些所獲得的分類樹，能被轉換成規則集合，並匯入知識庫中，幫助企業決策的制定和日常的運作。 此外，在兩個所提的知識整合方法中，第一個方法，能自動結合多重的知識來源成為一個整合的知識，並可匯入知識庫中，但是此方法只考慮到單一時間點的整合。第二個方法則是可以解決不同時間點的知識整合問題。另外，本論文提出三個新的遺傳運算子，在演化過程中，可用來解決規則集合中有重複、包含和衝突等常見的問題，因而可以產生較精簡及一致性高的分類規則。最後，本論文採用信用卡資料及乳癌資料來驗證所提方法的可行性，實驗結果皆獲得良好的成效。 / In today’s business environment, many enterprises make efforts in managing and applying organizational knowledge to sustain their core competence and create competitive advantage. The effective management of organizational knowledge can reduce the time and cost of solving problems, improve organizational performance, and increase organizational learning as well as innovative competence. Moreover, due to the need to accumulate knowledge resources, many enterprises have devoted to developing their knowledge repositories. These repositories store organizational and individual knowledge that are used to increase overall organization efficiency, support daily operations, and implement business strategies. Knowledge management (KM) is the modern paradigm for effective management of organizational knowledge to improve organizational performance. The intent of KM is to emphasize knowledge flows and the main process of acquisition, integration, storage/categorization, dissemination, and application. Furthermore, extant information technologies can provide a way of enabling more effective knowledge management. One of the important issues in knowledge management is knowledge acquisition. It is an extra burden for knowledge workers to contribute their knowledge into repositories, such that designing an effective method for abating an extra burden to automatically generate organizational knowledge will play a critical role in knowledge management. A second rather important issue in knowledge management is knowledge integration from different knowledge sources. Designing a knowledge-integration method to combine multiple knowledge sources will gradually become a necessity for enterprises. Classification problems, such as financial prediction and disease diagnosis, are often encountered in many applications. In the past, classification trees were often generated by decision-tree methods and commonly used to solve classification problems. In this dissertation, we propose two GP-based knowledge-acquisition methods and two GP-based knowledge-integration methods to support knowledge acquisition and knowledge integration respectively in the knowledge management processes for classification tasks. In the two proposed knowledge-acquisition methods, the first one is fast and easy to find the desired classification tree. It may, however, generate a complicated classification tree. The second method then further modifies the first method and produces a more concise and applicatory classification tree than the first one. The classification tree obtained can be transferred into a rule set, which can then be fed into a knowledge base to support decision making and facilitate daily operations. Furthermore, in the two proposed knowledge-integration methods, the former method can automatically combine multiple knowledge sources into one integrated knowledge base; nevertheless, it focuses on a single time point to deal with such knowledge-integration problems. The latter method then extends the former one to handle integrating situations properly with different time points. Additionally, three new genetic operators are designed in the evolving process to remove redundancy, subsumption and contradiction, thus producing more concise and consistent classification rules than those without using them. Finally, the proposed methods are applied to credit card data and breast cancer data for evaluating their effectiveness. They are also compared with several well-known classification methods. The experimental results show the good performance and feasibility of the proposed approaches.

Approved for public release, distribution is unlimited / The United States Navy requires radical and innovative ways to model and design multifunction phased array radars. This thesis puts forth the concept that Genetic Algorithms, computer simulations that mirror the natural selection process to develop creative solutions to complex problems, would be extremely well suited in this application. The capability of a Genetic Algorithm to predict adequately the behavior of an array antenna with randomly located elements was verified with expected results through the design, construction, development and evaluation of a test-bed array. The test-bed array was constructed of commercially available components, including a unique and innovative application of a quadrature modulator microchip used in commercial communications applications. Corroboration of predicted beam patterns from both Genetic Algorithm and Method of Moments calculations was achieved in anechoic chamber measurements conducted with the test-bed array. Both H-plane and E-plane data runs were made with several phase steered beams. In all cases the measured data agreed with that predicted from both modeling programs. Although time limited experiments to beam forming and steering with phase shifting, the test-bed array is fully capable of beam forming and steering though both phase shifting and amplitude tapering. / Outstanding Thesis / Lieutenant Commander, United States Navy

[pt] Esta Pesquisa tem por objetivo utilizar a Regressão Simbólica por Programação Genética para encontrar uma equação analítica para a fronteira de exercício ótima (ou curva de gatilho) de uma opção sobre um ativo do qual o preço tem um comportamento simulado pelo processo estocástico conhecido como processo de reversão à média (PRM). Para o cálculo do valor de uma opção desde de sua aquisição até sua maturação, normalmente faz-se o uso do cálculo da fronteira de exercício ótimo. Esta curva separa ao longo do tempo a decisão de exercer ou não a opção. Sabendo-se que já existem soluções analíticas para calcular a fronteira de exercício ótimo quando o preço do ativo segue um Movimento Geométrico Browniano, e que tal solução genérica ainda não foi encontrada para o PRM, neste trabalho, foi proposto o uso da Programação Genética (PG) para encontrar tal solução analítica. A Programação Genética utilizou um conjunto de amostras de curvas de exercício ótimo parametrizadas segundo a variação da volatilidade e da taxa de juros livre de risco, para encontrar uma função analítica para a fronteira de exercício ótima, obtendo-se resultados satisfatórios. / [en] This research intends on to use the Symbolic Regression by Genetic Programming to find an analytical equation that represents an Optimal Exercise Boundary for an option of an asset having its price behavior simulated by a stochastic process known as Mean Reversion Process (MRP). To calculate an option value since its acquisition until its maturity, normally is used to calculate the Optimal Exercise Boundary. This frontier separates along the time the decision to exercise the option or not. Knowing there already are analytical solutions used to calculate the Optimal Exercise Boundary when the asset price follows the Geometric Brownian Motion, and such general solution was not found yet to MRP, in this work, it was proposed the use of Genetic Programming to find such analytical solution. The Genetic Programming used an amount of samples from optimal exercise curves parameterized according the change in the volatility and risk free interest rate, to find an analytical function that represents Optimal Exercise Boundary, achieving satisfactory results.

Otimizar simultaneamente os interesses dos usuários, anunciantes e publicadores é um grande desafio na área de publicidade computacional. Mais precisamente, a ordenação de anúncios, ou ad ranking, desempenha um papel central nesse desafio. Por outro lado, nem mesmo as melhores fórmulas ou algoritmos de ordenação são capazes de manter seu status por um longo tempo em um ambiente que está em constante mudança. Neste trabalho, apresentamos uma análise orientada a dados que mostra a importância de combinar diferentes dimensões de publicidade computacional por meio de uma abordagem evolutiva para ordenação de anúncios afim de responder a mudanças de forma mais eficaz. Nós avaliamos as dimensões de valor comercial, desempenho histórico de cliques, interesses dos usuários e a similaridade textual entre o anúncio e a página. Nessa avaliação, nós averiguamos o desempenho e a correlação das diferentes dimensões. Como consequência, nós desenvolvemos uma abordagem evolucionária para combinar essas dimensões. Essa abordagem é composta por três partes: um repositório de configurações para facilitar a implantação e avaliação de experimentos de ordenação; um componente evolucionário de avaliação orientado a dados; e um motor de programação genética para evoluir fórmulas de ordenação de anúncios. Nossa abordagem foi implementada com sucesso em um sistema real de publicidade computacional responsável por processar mais de quatorze bilhões de requisições de anúncio por mês. De acordo com nossos resultados, essas dimensões se complementam e nenhuma delas deve ser neglicenciada. Além disso, nós mostramos que a combinação evolucionária dessas dimensões não só é capaz de superar cada uma individualmente, como também conseguiu alcançar melhores resultados do que métodos estáticos de ordenação de anúncios. / Simultaneous optimization of users, advertisers and publishers\' interests has been a formidable challenge in online advertising. More concretely, ranking of advertising, or more simply ad ranking, has a central role in this challenge. However, even the best ranking formula or algorithm cannot withstand the ever-changing environment of online advertising for a long time. In this work, we present a data-driven analysis that shows the importance of combining different aspects of online advertising through an evolutionary approach for ad ranking in order to effectively respond to changes. We evaluated aspects ranging from bid values and previous click performance to user behavior and interests, including the textual similarity between ad and page. In this evaluation, we assessed commercial performance along with the correlation between different aspects. Therefore, we proposed an evolutionary approach for combining these aspects. This approach was composed of three parts: a configuration repository to facilitate deployment and evaluation of ranking experiments; an evolutionary data-based evaluation component; and a genetic programming engine to evolve ad ranking formulae. Our approach was successfully implemented in a real online advertising system that processes more than fourteen billion ad requests per month. According to our results, these aspects complement each other and none of them should be neglected. Moreover, we showed that the evolutionary combination of these aspects not only outperformed each of them individually, but was also able to achieve better overall results than static ad ranking methods.

Cette thèse porte sur le traitement du langage naturel et l'exploration de texte, à l'intersection de l'apprentissage automatique et de la statistique. Nous nous intéressons plus particulièrement aux schémas de pondération des termes (SPT) dans le contexte de l'apprentissage supervisé et en particulier à la classification de texte. Dans la classification de texte, la tâche de classification multi-étiquettes a suscité beaucoup d'intérêt ces dernières années. La classification multi-étiquettes à partir de données textuelles peut être trouvée dans de nombreuses applications modernes telles que la classification de nouvelles où la tâche est de trouver les catégories auxquelles appartient un article de presse en fonction de son contenu textuel (par exemple, politique, Moyen-Orient, pétrole), la classification du genre musical (par exemple, jazz, pop, oldies, pop traditionnelle) en se basant sur les commentaires des clients, la classification des films (par exemple, action, crime, drame), la classification des produits (par exemple, électronique, ordinateur, accessoires). La plupart des algorithmes d'apprentissage ne conviennent qu'aux problèmes de classification binaire. Par conséquent, les tâches de classification multi-étiquettes sont généralement transformées en plusieurs tâches binaires à label unique. Cependant, cette transformation introduit plusieurs problèmes. Premièrement, les distributions des termes ne sont considérés qu'en matière de la catégorie positive et de la catégorie négative (c'est-à-dire que les informations sur les corrélations entre les termes et les catégories sont perdues). Deuxièmement, il n'envisage aucune dépendance vis-à-vis des étiquettes (c'est-à-dire que les informations sur les corrélations existantes entre les classes sont perdues). Enfin, puisque toutes les catégories sauf une sont regroupées dans une seule catégories (la catégorie négative), les tâches nouvellement créées sont déséquilibrées. Ces informations sont couramment utilisées par les SPT supervisés pour améliorer l'efficacité du système de classification. Ainsi, après avoir présenté le processus de classification de texte multi-étiquettes, et plus particulièrement le SPT, nous effectuons une comparaison empirique de ces méthodes appliquées à la tâche de classification de texte multi-étiquette. Nous constatons que la supériorité des méthodes supervisées sur les méthodes non supervisées n'est toujours pas claire. Nous montrons ensuite que ces méthodes ne sont pas totalement adaptées au problème de la classification multi-étiquettes et qu'elles ignorent beaucoup d'informations statistiques qui pourraient être utilisées pour améliorer les résultats de la classification. Nous proposons donc un nouvel SPT basé sur le gain d'information. Cette nouvelle méthode prend en compte la distribution des termes, non seulement en ce qui concerne la catégorie positive et la catégorie négative, mais également en rapport avec toutes les autres catégories. Enfin, dans le but de trouver des SPT spécialisés qui résolvent également le problème des tâches déséquilibrées, nous avons étudié les avantages de l'utilisation de la programmation génétique pour générer des SPT pour la tâche de classification de texte. Contrairement aux études précédentes, nous générons des formules en combinant des informations statistiques à un niveau microscopique (par exemple, le nombre de documents contenant un terme spécifique) au lieu d'utiliser des SPT complets. De plus, nous utilisons des informations catégoriques telles que (par exemple, le nombre de catégories dans lesquelles un terme apparaît). Des expériences sont effectuées pour mesurer l'impact de ces méthodes sur les performances du modèle. Nous montrons à travers ces expériences que les résultats sont positifs. / This thesis deals with natural language processing and text mining, at the intersection of machine learning and statistics. We are particularly interested in Term Weighting Schemes (TWS) in the context of supervised learning and specifically the Text Classification (TC) task. In TC, the multi-label classification task has gained a lot of interest in recent years. Multi-label classification from textual data may be found in many modern applications such as news classification where the task is to find the categories that a newswire story belongs to (e.g., politics, middle east, oil), based on its textual content, music genre classification (e.g., jazz, pop, oldies, traditional pop) based on customer reviews, film classification (e.g. action, crime, drama), product classification (e.g. Electronics, Computers, Accessories). Traditional classification algorithms are generally binary classifiers, and they are not suited for the multi-label classification. The multi-label classification task is, therefore, transformed into multiple single-label binary tasks. However, this transformation introduces several issues. First, terms distributions are only considered in relevance to the positive and the negative categories (i.e., information on the correlations between terms and categories is lost). Second, it fails to consider any label dependency (i.e., information on existing correlations between classes is lost). Finally, since all categories but one are grouped into one category (the negative category), the newly created tasks are imbalanced. This information is commonly used by supervised TWS to improve the effectiveness of the classification system. Hence, after presenting the process of multi-label text classification, and more particularly the TWS, we make an empirical comparison of these methods applied to the multi-label text classification task. We find that the superiority of the supervised methods over the unsupervised methods is still not clear. We show then that these methods are not fully adapted to the multi-label classification problem and they ignore much statistical information that coul be used to improve the classification results. Thus, we propose a new TWS based on information gain. This new method takes into consideration the term distribution, not only regarding the positive and the negative categories but also in relevance to all classes. Finally, aiming at finding specialized TWS that also solve the issue of imbalanced tasks, we studied the benefits of using genetic programming for generating TWS for the text classification task. Unlike previous studies, we generate formulas by combining statistical information at a microscopic level (e.g., the number of documents that contain a specific term) instead of using complete TWS. Furthermore, we make use of categorical information such as (e.g., the number of categories where a term occurs). Experiments are made to measure the impact of these methods on the performance of the model. We show through these experiments that the results are positive.

Un système est défini par les entités et leurs interrelations dans un environnement qui est déterminé par une limite arbitraire. Les systèmes complexes présentent un comportement émergent sans un contrôleur central. Les systèmes concrets désignent ceux qui sont observables dans la réalité. Un modèle nous permet de comprendre, de contrôler et de prédire le comportement du système. Un modèle différentiel à partir d'un système pourrait être compris comme une sorte de loi physique sous-jacent représenté par l'un ou d'un ensemble d'équations différentielles. Ce travail vise à étudier et mettre en œuvre des méthodes pour effectuer la modélisation des systèmes automatisée par l'ordinateur. Cette thèse pourrait être divisée en trois étapes principales, ainsi: (1) le développement d'un solveur numérique automatisé par l'ordinateur pour les équations différentielles linéaires, partielles ou ordinaires, sur la base de la formulation de matrice pour une personnalisation propre de la méthode Ritz-Galerkin; (2) la proposition d'un schème de score d'adaptation qui bénéficie du solveur numérique développé pour guider l'évolution des modèles différentiels pour les systèmes complexes concrets; (3) une implémentation préliminaire d'une application de programmation génétique pour effectuer la modélisation des systèmes automatisée par l'ordinateur. Dans la première étape, il est montré comment le solveur proposé utilise les polynômes de Jacobi orthogonaux comme base complète pour la méthode de Galerkin et comment le solveur traite des conditions auxiliaires de plusieurs types. Solutions à approximations polynomiales sont ensuite réalisés pour plusieurs types des équations différentielles partielles linéaires, y compris les problèmes hyperboliques, paraboliques et elliptiques. Dans la deuxième étape, le schème de score d'adaptation proposé est conçu pour exploiter certaines caractéristiques du solveur proposé et d'effectuer l'approximation polynômiale par morceaux afin d'évaluer les individus différentiels à partir d'une population fournie par l'algorithme évolutionnaire. Enfin, une mise en œuvre préliminaire d'une application GP est présentée et certaines questions sont discutées afin de permettre une meilleure compréhension de la modélisation des systèmes automatisée par l'ordinateur. Indications pour certains sujets prometteurs pour la continuation de futures recherches sont également abordées dans ce travail, y compris la façon d'étendre ce travail à certaines classes d'équations différentielles partielles non-linéaires. / A system is defined by its entities and their interrelations in an environment which is determined by an arbitrary boundary. Complex systems exhibit emergent behaviour without a central controller. Concrete systems designate the ones observable in reality. A model allows us to understand, to control and to predict behaviour of the system. A differential model from a system could be understood as some sort of underlying physical law depicted by either one or a set of differential equations. This work aims to investigate and implement methods to perform computer-automated system modelling. This thesis could be divided into three main stages: (1) developments of a computer-automated numerical solver for linear differential equations, partial or ordinary, based on the matrix formulation for an own customization of the Ritz-Galerkin method; (2) proposition of a fitness evaluation scheme which benefits from the developed numerical solver to guide evolution of differential models for concrete complex systems; (3) preliminary implementations of a genetic programming application to perform computer-automated system modelling. In the first stage, it is shown how the proposed solver uses Jacobi orthogonal polynomials as a complete basis for the Galerkin method and how the solver deals with auxiliary conditions of several types. Polynomial approximate solutions are achieved for several types of linear partial differential equations, including hyperbolic, parabolic and elliptic problems. In the second stage, the proposed fitness evaluation scheme is developed to exploit some characteristics from the proposed solver and to perform piecewise polynomial approximations in order to evaluate differential individuals from a given evolutionary algorithm population. Finally, a preliminary implementation of a genetic programming application is presented and some issues are discussed to enable a better understanding of computer-automated system modelling. Indications for some promising subjects for future continuation researches are also addressed here, as how to expand this work to some classes of non-linear partial differential equations.

本論文是承續近來〝有限理性總體經濟學〞發展下之一支研究。有關有限理性的定義，在本研究中乃是以Sargent(1993)及Leijonhufvud(1993)為根據。Sargent(1993)認為：經濟學家在建立模型時，要怎麼樣去塑造其模型中的決策者的預期及學習呢？為了在精神上求一致起見，不應將模型中的決策者想成比經濟學家本人更聰明或更無知。有關這兩個角色應一致的要求，似乎便成了有限理性總體經濟學中相當關鍵的磐石。有關預期與學習形成的部份在計量經濟學上，又可大致分為兩個階段。在第一階段中，是以統計決策理論為主所建構的預期與學習過程，這類型的預期是奠基於以機率模型為主的學習過程。此類學習過程可以說是1980年代以來，理性預期學習過程的主要架構。使用這種學習模型需對決策者在所擁有的資訊上，做較強的限制。而第二階段的學習模式是要減輕模型中對決策人在資訊上的負荷，即將第一階段機率模型的學習擴充至非機率模型的學習。而幾乎所有學習上的問題，都可以視為一個尋找的問題，模型選擇是尋找模型，參數估計是尋找參數。在模型的設定上，以往我們處理的程序是：假設模型為....，則我們可以....。對於模型的選定並沒有嚴格的判定標準可供依循。然而遺傳規畫不但對模型的設立，提供了一個良好的典範，而且對如何尋找模型，提供了一個一般性的尋找模式。模型的選取，應是先經由尋找的過程而得到的，而非憑空自上帝的手中取得。因此，就如何建立起尋找的方式，其較模型的選擇更為基本且更為重要。遺傳規畫運作之初，並沒有包含先驗的知識，初始的模型是經由隨機創造而得。在演化的過程中，模型逐漸地有了系統（型態）的出現。這種尋找的過程，既不偏向隨機也不偏向系統，在隨機與系統中，取得了一個完美的平衡點。在遺傳規畫運作下，要選擇何種模型，將視實驗者的時間成本而定。換句話說，即遺傳規畫提供了實驗者到目前為止最好的模型，是否該花更多時間以取得〝較精確〞的模型，將由實驗者自行決定。在此情況下，我們在模型的選擇上，有了一個較為適當的判定基準：模型的大體輪廓將是藉由進化的方式取得，不是經由天外神來之筆而誕生。在模型精確度的選擇上，將由個人的時間成本來定奪。就在這層意義上來說，此種選擇的模式比較符合〝人性〞，亦與經濟學的精神相符合。本論文的目的便是要了解遺傳規畫在實際運作上的一些特性，以及該如何正確地使用它才能得到最大的功效，以期望它能成為我們在處理有限理性總體經濟學上的一個重要工具。

Els paradigmes cel·lulars, com les xarxes neuronals cel·lulars (CNN, en anglès) i els autòmats cel·lulars (CA, en anglès), són una eina excel·lent de càlcul, al ser equivalents a una màquina universal de Turing. La introducció de la màquina universal CNN (CNN-UM, en anglès) ha permès desenvolupar hardware, el nucli computacional del qual funciona segons la filosofia cel·lular; aquest hardware ha trobat aplicació en diversos camps al llarg de la darrera dècada. Malgrat això, encara hi ha moltes preguntes a obertes sobre com definir els algoritmes d'una CNN-UM i com estudiar la dinàmica dels autòmats cel·lulars. En aquesta tesis es tracten els dos problemes: primer, es demostra que es possible acotar l'espai dels algoritmes per a la CNN-UM i explorar-lo gràcies a les tècniques genètiques; i segon, s'expliquen els fonaments de l'estudi dels CA per mitjà de la dinàmica no lineal (segons la definició de Chua) i s'il·lustra com aquesta tècnica ha permès trobar resultats innovadors. / Los paradigmas celulares, como las redes neuronales celulares (CNN, eninglés) y los autómatas celulares (CA, en inglés), son una excelenteherramienta de cálculo, al ser equivalentes a una maquina universal deTuring. La introducción de la maquina universal CNN (CNN-UM, eninglés) ha permitido desarrollar hardware cuyo núcleo computacionalfunciona según la filosofía celular; dicho hardware ha encontradoaplicación en varios campos a lo largo de la ultima década. Sinembargo, hay aun muchas preguntas abiertas sobre como definir losalgoritmos de una CNN-UM y como estudiar la dinámica de los autómatascelular. En esta tesis se tratan ambos problemas: primero se demuestraque es posible acotar el espacio de los algoritmos para la CNN-UM yexplorarlo gracias a técnicas genéticas; segundo, se explican losfundamentos del estudio de los CA por medio de la dinámica no lineal(según la definición de Chua) y se ilustra como esta técnica hapermitido encontrar resultados novedosos. / Cellular paradigms, like Cellular Neural Networks (CNNs) and Cellular Automata (CA) are an excellent tool to perform computation, since they are equivalent to a Universal Turing machine. The introduction of the Cellular Neural Network - Universal Machine (CNN-UM) allowed us to develop hardware whose computational core works according to the principles of cellular paradigms; such a hardware has found application in a number of fields throughout the last decade. Nevertheless, there are still many open questions about how to define algorithms for a CNN-UM, and how to study the dynamics of Cellular Automata. In this dissertation both problems are tackled: first, we prove that it is possible to bound the space of all algorithms of CNN-UM and explore it through genetic techniques; second, we explain the fundamentals of the nonlinear perspective of CA (according to Chua's definition), and we illustrate how this technique has allowed us to find novel results.