Personalised placement in networked video

Personalised video can be achieved by inserting objects into a video play-out according to the viewer's profile. Content which has been authored and produced for general broadcast can take on additional commercial service features when personalised either for individual viewers or for groups of viewers participating in entertainment, training, gaming or informational activities. Although several scenarios and use-cases can be envisaged, we are focussed on the application of personalised product placement. Targeted advertising and product placement are currently garnering intense interest in the commercial networked media industries. Personalisation of product placement is a relevant and timely service for next generation online marketing and advertising and for many other revenue generating interactive services. This paper discusses the acquisition and insertion of media objects into a TV video play-out stream where the objects are determined by the profile of the viewer. The technology is based on MPEG-4 standards using object based video and MPEG-7 for metadata. No proprietary technology or protocol is proposed. To trade the objects into the video play-out, a Software-as-a-Service brokerage platform based on intelligent agent technology is adopted. Agencies, libraries and service providers are represented in a commercial negotiation to facilitate the contractual selection and usage of objects to be inserted into the video play-out.

Poker learner: modelação de jogadores através de data mining

Dissertação de mestrado integrado em Engenharia e Gestão de Sistemas de Informação

Building the future: the Brazilian university library in 2010

Information technology will affect academic activities as well as the nature of the high education sector. This sector besides the need to assimilate these technologies will need to attend the requisites of market globalization and, as consequence, all theses changes will be reflected in the university library. Prospectives impacts will affect the structure (emphasis in user services, outsourcing of several services), in the financing aspect (growing of consortia in order to reduce costs), in services (electronic reference, support to long distance education programs, intelligent agents) and in the clientele (attending the great demand por high education which implies a diversity of people).

A logic programming framework for possibilistic argumentation: formalization and logical properties

In the last decade defeasible argumentation frameworks have evolved to become a sound setting to formalize commonsense, qualitative reasoning. The logic programming paradigm has shown to be particularly useful for developing different argument-based frameworks on the basis of different variants of logic programming which incorporate defeasible rules. Most of such frameworks, however, are unable to deal with explicit uncertainty, nor with vague knowledge, as defeasibility is directly encoded in the object language. This paper presents Possibilistic Logic Programming (P-DeLP), a new logic programming language which combines features from argumentation theory and logic programming, incorporating as well the treatment of possibilistic uncertainty. Such features are formalized on the basis of PGL, a possibilistic logic based on G¨odel fuzzy logic. One of the applications of P-DeLP is providing an intelligent agent with non-monotonic, argumentative inference capabilities. In this paper we also provide a better understanding of such capabilities by defining two non-monotonic operators which model the expansion of a given program P by adding new weighed facts associated with argument conclusions and warranted literals, respectively. Different logical properties for the proposed operators are studied

Understanding deep architectures and the effect of unsupervised pre-training

Cette thèse porte sur une classe d'algorithmes d'apprentissage appelés architectures profondes. Il existe des résultats qui indiquent que les représentations peu profondes et locales ne sont pas suffisantes pour la modélisation des fonctions comportant plusieurs facteurs de variation. Nous sommes particulièrement intéressés par ce genre de données car nous espérons qu'un agent intelligent sera en mesure d'apprendre à les modéliser automatiquement; l'hypothèse est que les architectures profondes sont mieux adaptées pour les modéliser. Les travaux de Hinton (2006) furent une véritable percée, car l'idée d'utiliser un algorithme d'apprentissage non-supervisé, les machines de Boltzmann restreintes, pour l'initialisation des poids d'un réseau de neurones supervisé a été cruciale pour entraîner l'architecture profonde la plus populaire, soit les réseaux de neurones artificiels avec des poids totalement connectés. Cette idée a été reprise et reproduite avec succès dans plusieurs contextes et avec une variété de modèles. Dans le cadre de cette thèse, nous considérons les architectures profondes comme des biais inductifs. Ces biais sont représentés non seulement par les modèles eux-mêmes, mais aussi par les méthodes d'entraînement qui sont souvent utilisés en conjonction avec ceux-ci. Nous désirons définir les raisons pour lesquelles cette classe de fonctions généralise bien, les situations auxquelles ces fonctions pourront être appliquées, ainsi que les descriptions qualitatives de telles fonctions. L'objectif de cette thèse est d'obtenir une meilleure compréhension du succès des architectures profondes. Dans le premier article, nous testons la concordance entre nos intuitions---que les réseaux profonds sont nécessaires pour mieux apprendre avec des données comportant plusieurs facteurs de variation---et les résultats empiriques. Le second article est une étude approfondie de la question: pourquoi l'apprentissage non-supervisé aide à mieux généraliser dans un réseau profond? Nous explorons et évaluons plusieurs hypothèses tentant d'élucider le fonctionnement de ces modèles. Finalement, le troisième article cherche à définir de façon qualitative les fonctions modélisées par un réseau profond. Ces visualisations facilitent l'interprétation des représentations et invariances modélisées par une architecture profonde.

Intelligence by Design: Principles of Modularity and Coordination for Engineerin

All intelligence relies on search --- for example, the search for an intelligent agent's next action. Search is only likely to succeed in resource-bounded agents if they have already been biased towards finding the right answer. In artificial agents, the primary source of bias is engineering. This dissertation describes an approach, Behavior-Oriented Design (BOD) for engineering complex agents. A complex agent is one that must arbitrate between potentially conflicting goals or behaviors. Behavior-oriented design builds on work in behavior-based and hybrid architectures for agents, and the object oriented approach to software engineering. The primary contributions of this dissertation are: 1.The BOD architecture: a modular architecture with each module providing specialized representations to facilitate learning. This includes one pre-specified module and representation for action selection or behavior arbitration. The specialized representation underlying BOD action selection is Parallel-rooted, Ordered, Slip-stack Hierarchical (POSH) reactive plans. 2.The BOD development process: an iterative process that alternately scales the agent's capabilities then optimizes the agent for simplicity, exploiting tradeoffs between the component representations. This ongoing process for controlling complexity not only provides bias for the behaving agent, but also facilitates its maintenance and extendibility. The secondary contributions of this dissertation include two implementations of POSH action selection, a procedure for identifying useful idioms in agent architectures and using them to distribute knowledge across agent paradigms, several examples of applying BOD idioms to established architectures, an analysis and comparison of the attributes and design trends of a large number of agent architectures, a comparison of biological (particularly mammalian) intelligence to artificial agent architectures, a novel model of primate transitive inference, and many other examples of BOD agents and BOD development.

Reinforcement Learning by Policy Search

One objective of artificial intelligence is to model the behavior of an intelligent agent interacting with its environment. The environment's transformations can be modeled as a Markov chain, whose state is partially observable to the agent and affected by its actions; such processes are known as partially observable Markov decision processes (POMDPs). While the environment's dynamics are assumed to obey certain rules, the agent does not know them and must learn. In this dissertation we focus on the agent's adaptation as captured by the reinforcement learning framework. This means learning a policy---a mapping of observations into actions---based on feedback from the environment. The learning can be viewed as browsing a set of policies while evaluating them by trial through interaction with the environment. The set of policies is constrained by the architecture of the agent's controller. POMDPs require a controller to have a memory. We investigate controllers with memory, including controllers with external memory, finite state controllers and distributed controllers for multi-agent systems. For these various controllers we work out the details of the algorithms which learn by ascending the gradient of expected cumulative reinforcement. Building on statistical learning theory and experiment design theory, a policy evaluation algorithm is developed for the case of experience re-use. We address the question of sufficient experience for uniform convergence of policy evaluation and obtain sample complexity bounds for various estimators. Finally, we demonstrate the performance of the proposed algorithms on several domains, the most complex of which is simulated adaptive packet routing in a telecommunication network.

Multidimensional Adaptations for Open Learning Management Systems

Our work is focused on alleviating the workload for designers of adaptive courses on the complexity task of authoring adaptive learning designs adjusted to specific user characteristics and the user context. We propose an adaptation platform that consists in a set of intelligent agents where each agent carries out an independent adaptation task. The agents apply machine learning techniques to support the user modelling for the adaptation process

Algoritmo Q-learning como estratégia de exploração e/ou explotação para metaheurísticas GRASP e algoritmo genético

Techniques of optimization known as metaheuristics have achieved success in the resolution of many problems classified as NP-Hard. These methods use non deterministic approaches that reach very good solutions which, however, don t guarantee the determination of the global optimum. Beyond the inherent difficulties related to the complexity that characterizes the optimization problems, the metaheuristics still face the dilemma of xploration/exploitation, which consists of choosing between a greedy search and a wider exploration of the solution space. A way to guide such algorithms during the searching of better solutions is supplying them with more knowledge of the problem through the use of a intelligent agent, able to recognize promising regions and also identify when they should diversify the direction of the search. This way, this work proposes the use of Reinforcement Learning technique - Q-learning Algorithm - as exploration/exploitation strategy for the metaheuristics GRASP (Greedy Randomized Adaptive Search Procedure) and Genetic Algorithm. The GRASP metaheuristic uses Q-learning instead of the traditional greedy-random algorithm in the construction phase. This replacement has the purpose of improving the quality of the initial solutions that are used in the local search phase of the GRASP, and also provides for the metaheuristic an adaptive memory mechanism that allows the reuse of good previous decisions and also avoids the repetition of bad decisions. In the Genetic Algorithm, the Q-learning algorithm was used to generate an initial population of high fitness, and after a determined number of generations, where the rate of diversity of the population is less than a certain limit L, it also was applied to supply one of the parents to be used in the genetic crossover operator. Another significant change in the hybrid genetic algorithm is the proposal of a mutually interactive cooperation process between the genetic operators and the Q-learning algorithm. In this interactive/cooperative process, the Q-learning algorithm receives an additional update in the matrix of Q-values based on the current best solution of the Genetic Algorithm. The computational experiments presented in this thesis compares the results obtained with the implementation of traditional versions of GRASP metaheuristic and Genetic Algorithm, with those obtained using the proposed hybrid methods. Both algorithms had been applied successfully to the symmetrical Traveling Salesman Problem, which was modeled as a Markov decision process

Projeto e construção de um robô móvel AGV/ROV não-holonômico com habilidade para navegação autônoma do tipo wall-following

Esta dissertação de mestrado apresenta o projeto e a construção de um robô móvel terrestre denominado LOGBOT, com tração de movimento do tipo diferencial – com duas rodas motoras e uma roda livre para manter a estabilidade de sua estrutura em relação à superfície. O controle do robô dispõe dos modos de telemetria e autônomo. No modo de controle por telemetria (ROV), a comunicação do robô com a estação de controle é feita por radiofreqüência a uma distância de até um quilometro em ambientes externos, e até cem metros em ambientes internos. No modo de controle autônomo (AGV), o robô tem habilidade para navegar em ambientes internos e desconhecidos usando sempre a parede à sua esquerda como referência para a trajetória de seu movimento. A seqüência de movimentos para execução da trajetória é enviada para a estação de controle que realiza análises de desempenho do robô. Para executar suas tarefas no modo autônomo, a programação do robô conta com um agente inteligente reativo, que detecta características do ambiente (obstáculos, final de paredes, etc.) e decide sobre qual atitude deve ser executada pelo robô, com objetivo de contornar os obstáculos e controlar a velocidade de suas rodas. Os problemas de erro odométrico e suas correções com base no uso de informações sensoriais externas são devidamente tratados. Técnicas de controle hierárquico do robô como um todo e controle em malha fechada da velocidade das rodas do robô são usadas. Os resultados mostraram que o robô móvel LOGBOT é capaz de navegar, com estabilidade e precisão, em ambientes internos no formato de um corredor (wall following).

Sistema multiagente web semântico para gestão de conteúdos educacionais

Pós-graduação em Ciência da Computação - IBILCE

A framework for goal-oriented discovery of resources in the RESTful architecture

One of the challenges facing the current web is the efficient use of all the available information. The Web 2.0 phenomenon has favored the creation of contents by average users, and thus the amount of information that can be found for diverse topics has grown exponentially in the last years. Initiatives such as linked data are helping to build the Semantic Web, in which a set of standards are proposed for the exchange of data among heterogeneous systems. However, these standards are sometimes not used, and there are still plenty of websites that require naive techniques to discover their contents and services. This paper proposes an integrated framework for content and service discovery and extraction. The framework is divided into several layers where the discovery of contents and services is made in a representational stateless transfer system such as the web. It employs several web mining techniques as well as feature-oriented modeling for the discovery of cross-cutting features in web resources. The framework is used in a scenario of electronic newspapers. An intelligent agent crawls the web for related news, and uses services and visits links automatically according to its goal. This scenario illustrates how the discovery is made at different levels and how the use of semantics helps implement an agent that performs high-level tasks.