Improving Collaborative Filtering Based Recommender Systems Using Pareto Dominance

Los sistemas de recomendación son un tipo de solución al problema de sobrecarga de información que sufren los usuarios de los sitios web en los que se pueden votar ciertos artículos. El sistema de recomendación de filtrado colaborativo es considerado como el método con más éxito debido a que sus recomendaciones se hacen basándose en los votos de usuarios similares a un usuario activo. Sin embargo, el método de filtrado de colaboración tradicional selecciona usuarios insuficientemente representativos como vecinos de cada usuario activo. Esto significa que las recomendaciones hechas a posteriori no son lo suficientemente precisas. El método propuesto en esta tesis realiza un pre-filtrado del proceso, mediante el uso de dominancia de Pareto, que elimina los usuarios menos representativos del proceso de selección k-vecino y mantiene los más prometedores. Los resultados de los experimentos realizados en MovieLens y Netflix muestran una mejora significativa en todas las medidas de calidad estudiadas en la aplicación del método propuesto. ABSTRACTRecommender systems are a type of solution to the information overload problem suffered by users of websites on which they can rate certain items. The Collaborative Filtering Recommender System is considered to be the most successful approach as it make its recommendations based on votes of users similar to an active user. Nevertheless, the traditional collaborative filtering method selects insufficiently representative users as neighbors of each active user. This means that the recommendations made a posteriori are not precise enough. The method proposed in this thesis performs a pre-filtering process, by using Pareto dominance, which eliminates the less representative users from the k-neighbor selection process and keeps the most promising ones. The results from the experiments performed on Movielens and Netflix show a significant improvement in all the quality measures studied on applying the proposed method.

Pareto Dominance of Deferred Acceptance through Early Decision

An early decision market is governed by rules that allow each student to apply to (at most) one college and require the student to attend this college if admitted. This market is ubiquitous in college admissions in the United States. We model this market as an extensive-form game of perfect information and study a refinement of subgame perfect equilibrium (SPE) that induces undominated Nash equilibria in every subgame (SPUE). Our main result shows that this game can be used to define a decentralized matching mechanism that weakly Pareto dominates student-proposing deferred acceptance.

Pareto Dominance of Deferred Acceptance through Early Decision

An early decision market is governed by rules that allow each student to apply to (at most) one college and require the student to attend this college if admitted. This market is ubiquitous in college admissions in the United States. We model this market as an extensive-form game of perfect information and study a refinement of subgame perfect equilibrium (SPE) that induces undominated Nash equilibria in every subgame (SPUE). Our main result shows that this game can be used to define a decentralized matching mechanism that weakly Pareto dominates student-proposing deferred acceptance.

Ranking-Dominance Relation for (Evolutionary) Multi-Objective Optimization

An alternative relation to Pareto-dominance relation is proposed. The new relation is based on ranking a set of solutions according to each separate objective and an aggregation function to calculate a scalar fitness value for each solution. The relation is called as ranking-dominance and it tries to tackle the curse of dimensionality commonly observedin evolutionary multi-objective optimization. Ranking-dominance can beused to sort a set of solutions even for a large number of objectives when Pareto-dominance relation cannot distinguish solutions from one another anymore. This permits search to advance even with a large number of objectives. It is also shown that ranking-dominance does not violate Pareto-dominance. Results indicate that selection based on ranking-dominance is able to advance search towards the Pareto-front in some cases, where selection based on Pareto-dominance stagnates. However, in some cases it is also possible that search does not proceed into direction of Pareto-front because the ranking-dominance relation permits deterioration of individual objectives. Results also show that when the number of objectives increases, selection based on just Pareto-dominance without diversity maintenance is able to advance search better than with diversity maintenance. Therefore, diversity maintenance is connive at the curse of dimensionality.

Direct multisearch for multiobjective optimization

In practical applications of optimization it is common to have several conflicting objective functions to optimize. Frequently, these functions are subject to noise or can be of black-box type, preventing the use of derivative-based techniques. We propose a novel multiobjective derivative-free methodology, calling it direct multisearch (DMS), which does not aggregate any of the objective functions. Our framework is inspired by the search/poll paradigm of direct-search methods of directional type and uses the concept of Pareto dominance to maintain a list of nondominated points (from which the new iterates or poll centers are chosen). The aim of our method is to generate as many points in the Pareto front as possible from the polling procedure itself, while keeping the whole framework general enough to accommodate other disseminating strategies, in particular, when using the (here also) optional search step. DMS generalizes to multiobjective optimization (MOO) all direct-search methods of directional type. We prove under the common assumptions used in direct search for single objective optimization that at least one limit point of the sequence of iterates generated by DMS lies in (a stationary form of) the Pareto front. However, extensive computational experience has shown that our methodology has an impressive capability of generating the whole Pareto front, even without using a search step. Two by-products of this paper are (i) the development of a collection of test problems for MOO and (ii) the extension of performance and data profiles to MOO, allowing a comparison of several solvers on a large set of test problems, in terms of their efficiency and robustness to determine Pareto fronts.

Node adaptation for global collocation with radial basis functions using direct multisearch for multiobjective optimization

Meshless methods are used for their capability of producing excellent solutions without requiring a mesh, avoiding mesh related problems encountered in other numerical methods, such as finite elements. However, node placement is still an open question, specially in strong form collocation meshless methods. The number of used nodes can have a big influence on matrix size and therefore produce ill-conditioned matrices. In order to optimize node position and number, a direct multisearch technique for multiobjective optimization is used to optimize node distribution in the global collocation method using radial basis functions. The optimization method is applied to the bending of isotropic simply supported plates. Using as a starting condition a uniformly distributed grid, results show that the method is capable of reducing the number of nodes in the grid without compromising the accuracy of the solution. (C) 2013 Elsevier Ltd. All rights reserved.

Paretian Quasi-Orders: Two Agents.

We characterize Paretian quasi-orders in the two-agent continuous case.

Um estudo algorítmico para otimização do plano de tratamento da radioterapia conformal

This work performs an algorithmic study of optimization of a conformal radiotherapy plan treatment. Initially we show: an overview about cancer, radiotherapy and the physics of interaction of ionizing radiation with matery. A proposal for optimization of a plan of treatment in radiotherapy is developed in a systematic way. We show the paradigm of multicriteria problem, the concept of Pareto optimum and Pareto dominance. A generic optimization model for radioterapic treatment is proposed. We construct the input of the model, estimate the dose given by the radiation using the dose matrix, and show the objective function for the model. The complexity of optimization models in radiotherapy treatment is typically NP which justifyis the use of heuristic methods. We propose three distinct methods: MOGA, MOSA e MOTS. The project of these three metaheuristic procedures is shown. For each procedures follows: a brief motivation, the algorithm itself and the method for tuning its parameters. The three method are applied to a concrete case and we confront their performances. Finally it is analyzed for each method: the quality of the Pareto sets, some solutions and the respective Pareto curves

Regularized model learning in EDAs for continuous and multi-objective optimization

Probabilistic modeling is the de�ning characteristic of estimation of distribution algorithms (EDAs) which determines their behavior and performance in optimization. Regularization is a well-known statistical technique used for obtaining an improved model by reducing the generalization error of estimation, especially in high-dimensional problems. `1-regularization is a type of this technique with the appealing variable selection property which results in sparse model estimations. In this thesis, we study the use of regularization techniques for model learning in EDAs. Several methods for regularized model estimation in continuous domains based on a Gaussian distribution assumption are presented, and analyzed from di�erent aspects when used for optimization in a high-dimensional setting, where the population size of EDA has a logarithmic scale with respect to the number of variables. The optimization results obtained for a number of continuous problems with an increasing number of variables show that the proposed EDA based on regularized model estimation performs a more robust optimization, and is able to achieve signi�cantly better results for larger dimensions than other Gaussian-based EDAs. We also propose a method for learning a marginally factorized Gaussian Markov random �eld model using regularization techniques and a clustering algorithm. The experimental results show notable optimization performance on continuous additively decomposable problems when using this model estimation method. Our study also covers multi-objective optimization and we propose joint probabilistic modeling of variables and objectives in EDAs based on Bayesian networks, speci�cally models inspired from multi-dimensional Bayesian network classi�ers. It is shown that with this approach to modeling, two new types of relationships are encoded in the estimated models in addition to the variable relationships captured in other EDAs: objectivevariable and objective-objective relationships. An extensive experimental study shows the e�ectiveness of this approach for multi- and many-objective optimization. With the proposed joint variable-objective modeling, in addition to the Pareto set approximation, the algorithm is also able to obtain an estimation of the multi-objective problem structure. Finally, the study of multi-objective optimization based on joint probabilistic modeling is extended to noisy domains, where the noise in objective values is represented by intervals. A new version of the Pareto dominance relation for ordering the solutions in these problems, namely �-degree Pareto dominance, is introduced and its properties are analyzed. We show that the ranking methods based on this dominance relation can result in competitive performance of EDAs with respect to the quality of the approximated Pareto sets. This dominance relation is then used together with a method for joint probabilistic modeling based on `1-regularization for multi-objective feature subset selection in classi�cation, where six di�erent measures of accuracy are considered as objectives with interval values. The individual assessment of the proposed joint probabilistic modeling and solution ranking methods on datasets with small-medium dimensionality, when using two di�erent Bayesian classi�ers, shows that comparable or better Pareto sets of feature subsets are approximated in comparison to standard methods.

Interval-based ranking in noisy evolutionary multiobjective optimization

As one of the most competitive approaches to multi-objective optimization, evolutionary algorithms have been shown to obtain very good results for many realworld multi-objective problems. One of the issues that can affect the performance of these algorithms is the uncertainty in the quality of the solutions which is usually represented with the noise in the objective values. Therefore, handling noisy objectives in evolutionary multi-objective optimization algorithms becomes very important and is gaining more attention in recent years. In this paper we present ?-degree Pareto dominance relation for ordering the solutions in multi-objective optimization when the values of the objective functions are given as intervals. Based on this dominance relation, we propose an adaptation of the non-dominated sorting algorithm for ranking the solutions. This ranking method is then used in a standardmulti-objective evolutionary algorithm and a recently proposed novel multi-objective estimation of distribution algorithm based on joint variable-objective probabilistic modeling, and applied to a set of multi-objective problems with different levels of independent noise. The experimental results show that the use of the proposed method for solution ranking allows to approximate Pareto sets which are considerably better than those obtained when using the dominance probability-based ranking method, which is one of the main methods for noise handling in multi-objective optimization.

Combining non-dominance, objective-order and spread metric to extend firefly algorithm to multi-objective optimization

In this paper, we propose an extension of the firefly algorithm (FA) to multi-objective optimization. FA is a swarm intelligence optimization algorithm inspired by the flashing behavior of fireflies at night that is capable of computing global solutions to continuous optimization problems. Our proposal relies on a fitness assignment scheme that gives lower fitness values to the positions of fireflies that correspond to non-dominated points with smaller aggregation of objective function distances to the minimum values. Furthermore, FA randomness is based on the spread metric to reduce the gaps between consecutive non-dominated solutions. The obtained results from the preliminary computational experiments show that our proposal gives a dense and well distributed approximated Pareto front with a large number of points.

Seed rain in areas with and without bamboo dominance within an urban fragment of the Atlantic Forest

Understanding the flow of diaspores is fundamental for determining plant population dynamics in a particular habitat, and a lack of seeds is a limiting factor in forest regeneration, especially in isolated forest fragments. Bamboo dominance affects forest structure and dynamics by suppressing or delaying the recruitment of and colonization by tree species as well as by inhibiting the survival and growth of adult trees. The goal of the present study was to determine whether dominance of the bamboo species Aulonemia aristulata (Döll) McClure in the forest understory influences species abundance and composition. We examined the seed rain at two noncontiguous sites (1.5 km apart) within an urban forest fragment, with and without bamboo dominance (BD+ and BD- areas, respectively). Sixty seed traps (0.5 m², with a 1-mm mesh) were set in the BD+ and BD- areas, and the seed rain was sampled from January to December 2007. Diaspores were classified according to dispersal syndrome, growth form and functional type of the species to which they belonged. There were significant differences between the two areas in terms of seed density, species diversity and dispersal syndrome. The BD+ area showed greater seed density and species diversity. In both areas, seed distribution was limited primarily by impaired dispersal. Bamboo dominance and low tree density resulted in fewer propagules in the seed rain. Our results suggest that low availability of seeds in the rain does not promote the maintenance of a degraded state, characterized by the presence of bamboo.

Estimativa da probabilidade do evento extremo de precipitação de janeiro de 2000 no Vale do Paraíba, baseada na distribuição generalizada de pareto

Um evento extremo de precipitação ocorreu na primeira semana do ano 2000, de 1º a 5 de janeiro, no Vale do Paraíba, parte leste do Estado de São Paulo, Brasil, causando enorme impacto socioeconômico, com mortes e destruição. Este trabalho estudou este evento em 10 estações meteorológicas selecionadas que foram consideradas como aquelas tendo dados mais homogêneos do Que outras estações na região. O modelo de distribuição generalizada de Pareto (DGP) para valores extremos de precipitação de 5 dias foi desenvolvido, individualmente para cada uma dessas estações. Na modelagem da DGP, foi adotada abordagem não-estacionaria considerando o ciclo anual e tendência de longo prazo como co-variaveis. Uma conclusão desta investigação é que as quantidades de precipitação acumulada durante os 5 dias do evento estudado podem ser classificadas como extremamente raras para a região, com probabilidade de ocorrência menor do que 1% para maioria das estações, e menor do que 0,1% em três estações.

Gravity-Induced Vacuum Dominance

has been widely believed that, except in very extreme situations, the influence of gravity on quantum fields should amount to just small, subdominant contributions. This view seemed to be endorsed by the seminal results obtained over the last decades in the context of renormalization of quantum fields in curved spacetimes. Here, however, we argue that this belief is false by showing that there exist well-behaved spacetime evolutions where the vacuum energy density of free quantum fields is forced, by the very same background spacetime, to become dominant over any classical energy-density component. By estimating the time scale for the vacuum energy density to become dominant, and therefore for back-reaction on the background spacetime to become important, we argue that this (infrared) vacuum dominance may bear unexpected astrophysical and cosmological implications.

Multiobjective Particle Swarm Approach for the Design of a Brushless DC Wheel Motor

The roots of swarm intelligence are deeply embedded in the biological study of self-organized behaviors in social insects. Particle swarm optimization (PSO) is one of the modern metaheuristics of swarm intelligence, which can be effectively used to solve nonlinear and non-continuous optimization problems. The basic principle of PSO algorithm is formed on the assumption that potential solutions (particles) will be flown through hyperspace with acceleration towards more optimum solutions. Each particle adjusts its flying according to the flying experiences of both itself and its companions using equations of position and velocity. During the process, the coordinates in hyperspace associated with its previous best fitness solution and the overall best value attained so far by other particles within the group are kept track and recorded in the memory. In recent years, PSO approaches have been successfully implemented to different problem domains with multiple objectives. In this paper, a multiobjective PSO approach, based on concepts of Pareto optimality, dominance, archiving external with elite particles and truncated Cauchy distribution, is proposed and applied in the design with the constraints presence of a brushless DC (Direct Current) wheel motor. Promising results in terms of convergence and spacing performance metrics indicate that the proposed multiobjective PSO scheme is capable of producing good solutions.