Implementing Policy Management through BDI

The requirement for Grid middleware to be largely transparent to individual users and at the same time act in accordance with their personal needs is a difficult challenge. In e-science scenarios, users cannot be repeatedly interrogated for each operational decision made when enacting experiments on the Grid. It is thus important to specify and enforce policies that enable the environment to be configured to take user preferences into account automatically. In particular, we need to consider the context in which these policies are applied, because decisions are based not only on the rules of the policy but also on the current state of the system. Consideration of context is explicitly addressed, in the agent perspective, when deciding how to balance the achievement of goals and reaction to the environment. One commonly-applied abstraction that balances reaction to multiple events with context-based reasoning in the way suggested by our requirements is the belief-desire-intention (BDI) architecture, which has proven successful in many applications. In this paper, we argue that BDI is an appropriate model for policy enforcement, and describe the application of BDI to policy enforcement in personalising Grid service discovery. We show how this has been implemented in the myGrid registry to provide bioinformaticians with control over the services returned to them by the service discovery process.

The Requirements of Using Provenance in e-Science Experiments

In e-Science experiments, it is vital to record the experimental process for later use such as in interpreting results, verifying that the correct process took place or tracing where data came from. The process that led to some data is called the provenance of that data, and a provenance architecture is the software architecture for a system that will provide the necessary functionality to record, store and use process documentation. However, there has been little principled analysis of what is actually required of a provenance architecture, so it is impossible to determine the functionality they would ideally support. In this paper, we present use cases for a provenance architecture from current experiments in biology, chemistry, physics and computer science, and analyse the use cases to determine the technical requirements of a generic, technology and application-independent architecture. We propose an architecture that meets these requirements and evaluate a preliminary implementation by attempting to realise two of the use cases.

Using an animated pedagogical agent to interact affectively with the student

This work proposes an animated pedagogical agent that has the role of providing emotional support to the student: motivating and encouraging him, making him believe in his self-ability, and promoting a positive mood in him, which fosters learning. This careful support of the agent, its affective tactics, is expressed through emotional behaviour and encouragement messages of the lifelike character. Due to human social tendency of anthropomorphising software, we believe that a software agent can accomplish this affective role. In order to choose the adequate affective tactics, the agent should also know the student’s emotions. The proposed agent recognises the student’s emotions: joy/distress, satisfaction/disappointment, anger/gratitude, and shame, from the student’s observable behaviour, i. e. his actions in the interface of the educational system. The inference of emotions is psychologically grounded on the cognitive theory of emotions. More specifically, we use the OCC model which is based on the cognitive approach of emotion and can be computationally implemented. Due to the dynamic nature of the student’s affective information, we adopted a BDI approach to implement the affective user model and the affective diagnosis. Besides, in our work we profit from the reasoning capacity of the BDI approach in order for the agent to deduce the student’s appraisal, which allows it to infer the student’s emotions. As a case study, the proposed agent is implemented as the Mediating Agent of MACES: an educational collaborative environment modelled as a multi-agent system and pedagogically based on the sociocultural theory of Vygotsky.

Aplicação de classificadores Bayesianos e regressão logística na análise de desempenho dos alunos de graduação

Este trabalho minera as informações coletadas no processo de vestibular entre 2009 e 2012 para o curso de graduação de administração de empresas da FGV-EAESP, para estimar classificadores capazes de calcular a probabilidade de um novo aluno ter bom desempenho. O processo de KDD (Knowledge Discovery in Database) desenvolvido por Fayyad et al. (1996a) é a base da metodologia adotada e os classificadores serão estimados utilizando duas ferramentas matemáticas. A primeira é a regressão logística, muito usada por instituições financeiras para avaliar se um cliente será capaz de honrar com seus pagamentos e a segunda é a rede Bayesiana, proveniente do campo de inteligência artificial. Este estudo mostre que os dois modelos possuem o mesmo poder discriminatório, gerando resultados semelhantes. Além disso, as informações que influenciam a probabilidade de o aluno ter bom desempenho são a sua idade no ano de ingresso, a quantidade de vezes que ele prestou vestibular da FGV/EAESP antes de ser aprovado, a região do Brasil de onde é proveniente e as notas das provas de matemática fase 01 e fase 02, inglês, ciências humanas e redação. Aparentemente o grau de formação dos pais e o grau de decisão do aluno em estudar na FGV/EAESP não influenciam nessa probabilidade.

A Learning Function for Parameter Reduction in Spiking Neural Networks with Radial Basis Function

Spiking neural networks - networks that encode information in the timing of spikes - are arising as a new approach in the artificial neural networks paradigm, emergent from cognitive science. One of these new models is the pulsed neural network with radial basis function, a network able to store information in the axonal propagation delay of neurons. Learning algorithms have been proposed to this model looking for mapping input pulses into output pulses. Recently, a new method was proposed to encode constant data into a temporal sequence of spikes, stimulating deeper studies in order to establish abilities and frontiers of this new approach. However, a well known problem of this kind of network is the high number of free parameters - more that 15 - to be properly configured or tuned in order to allow network convergence. This work presents for the first time a new learning function for this network training that allow the automatic configuration of one of the key network parameters: the synaptic weight decreasing factor.

Técnicas de inteligência artificial para a geração dinâmica de set points para uma coluna de destilação

Artificial Intelligence techniques are applied to improve performance of a simulated oil distillation system. The chosen system was a debutanizer column. At this process, the feed, which comes to the column, is segmented by heating. The lightest components become steams, by forming the LPG (Liquefied Petroleum Gas). The others components, C5+, continue liquid. In the composition of the LPG, ideally, we have only propane and butanes, but, in practice, there are contaminants, for example, pentanes. The objective of this work is to control pentane amount in LPG, by means of intelligent set points (SP s) determination for PID controllers that are present in original instrumentation (regulatory control) of the column. A fuzzy system will be responsible for adjusting the SP's, driven by the comparison between the molar fraction of the pentane present in the output of the plant (LPG) and the desired amount. However, the molar fraction of pentane is difficult to measure on-line, due to constraints such as: long intervals of measurement, high reliability and low cost. Therefore, an inference system was used, based on a multilayer neural network, to infer the pentane molar fraction through secondary variables of the column. Finally, the results shown that the proposed control system were able to control the value of pentane molar fraction under different operational situations

Separação cega de fontes lineares e não lineares usando algoritmo genético, redes neurais artificiais RBF e negentropia de Rényi como medida de independência

Conventional methods to solve the problem of blind source separation nonlinear, in general, using series of restrictions to obtain the solution, often leading to an imperfect separation of the original sources and high computational cost. In this paper, we propose an alternative measure of independence based on information theory and uses the tools of artificial intelligence to solve problems of blind source separation linear and nonlinear later. In the linear model applies genetic algorithms and Rényi of negentropy as a measure of independence to find a separation matrix from linear mixtures of signals using linear form of waves, audio and images. A comparison with two types of algorithms for Independent Component Analysis widespread in the literature. Subsequently, we use the same measure of independence, as the cost function in the genetic algorithm to recover source signals were mixed by nonlinear functions from an artificial neural network of radial base type. Genetic algorithms are powerful tools for global search, and therefore well suited for use in problems of blind source separation. Tests and analysis are through computer simulations

Modeling and analysis of artificial neural networks applied in operations research

Artificial neural networks are dynamic systems consisting of highly interconnected and parallel nonlinear processing elements. Systems based on artificial neural networks have high computational rates due to the use of a massive number of these computational elements. Neural networks with feedback connections provide a computing model capable of solving a rich class of optimization problems. In this paper, a modified Hopfield network is developed for solving problems related to operations research. The internal parameters of the network are obtained using the valid-subspace technique. Simulated examples are presented as an illustration of the proposed approach. Copyright (C) 2000 IFAC.

Inteligência artificial e teoria de resolução de problemas

Este artigo é uma tentativa de delinear as principais características da pesquisa numa nova área de estudos a chamada Inteligência Artificial (AI). Os itens 1 e 2 constituem um rápido histórico da AI e seus pressupostos básicos. O item 3 trata da teoria de resolução de problemas, desenvolvida por A. Newell e H. Simon. O item 4 procura mostrar a relevância da AI para a Filosofia, em especial para a filosofia da Mente e para a Teoria do Conhecimento.

Robots, intencionalidade e inteligência artificial

O artigo aborda problemas filosóficos relativos à natureza da intencionalidade e da representação mental. A primeira parte apresenta um breve histórico dos problemas, percorrendo rapidamente alguns episódios da filosofia clássica e da filosofia contemporânea. A segunda parte examina o Chinese Room Argument (Argumento do Quarto do Chinês) formulado por J. Searle. A terceira parte desenvolve alguns argumentos visando mostrar a inadequação do modelo funcionalista de mente na construção de robots. A conclusão (quarta parte) aponta algumas alternativas ao modelo funcionalista tradicional, como, por exemplo, o conexionismo.

Lógica BDI fuzzy

Intendding to understand how the human mind operates, some philosophers and psycologists began to study about rationality. Theories were built from those studies and nowadays that interest have been extended to many other areas such as computing engineering and computing science, but with a minimal distinction at its goal: to understand the mind operational proccess and apply it on agents modelling to become possible the implementation (of softwares or hardwares) with the agent-oriented paradigm where agents are able to deliberate their own plans of actions. In computing science, the sub-area of multiagents systems has progressed using several works concerning artificial intelligence, computational logic, distributed systems, games theory and even philosophy and psycology. This present work hopes to show how it can be get a logical formalisation extention of a rational agents architecture model called BDI (based in a philosophic Bratman s Theory) in which agents are capable to deliberate actions from its beliefs, desires and intentions. The formalisation of this model is called BDI logic and it is a modal logic (in general it is a branching time logic) with three access relations: B, D and I. And here, it will show two possible extentions that tranform BDI logic in a modal-fuzzy logic where the formulae and the access relations can be evaluated by values from the interval [0,1]

Uma Análise de métodos de distriubuição de atributos em comitês de classificadores

The objective of the researches in artificial intelligence is to qualify the computer to execute functions that are performed by humans using knowledge and reasoning. This work was developed in the area of machine learning, that it s the study branch of artificial intelligence, being related to the project and development of algorithms and techniques capable to allow the computational learning. The objective of this work is analyzing a feature selection method for ensemble systems. The proposed method is inserted into the filter approach of feature selection method, it s using the variance and Spearman correlation to rank the feature and using the reward and punishment strategies to measure the feature importance for the identification of the classes. For each ensemble, several different configuration were used, which varied from hybrid (homogeneous) to non-hybrid (heterogeneous) structures of ensemble. They were submitted to five combining methods (voting, sum, sum weight, multiLayer Perceptron and naïve Bayes) which were applied in six distinct database (real and artificial). The classifiers applied during the experiments were k- nearest neighbor, multiLayer Perceptron, naïve Bayes and decision tree. Finally, the performance of ensemble was analyzed comparatively, using none feature selection method, using a filter approach (original) feature selection method and the proposed method. To do this comparison, a statistical test was applied, which demonstrate that there was a significant improvement in the precision of the ensembles