Parallel implementation of an adaptive generalized predictive control algorithm

The Adaptive Generalized Predictive Control (GPC) algorithm can be speeded up using parallel processing. Since the GPC algorithm needs to be fed with knowledge of the plant transfer function, the parallelization of a standard Recursive Least Squares (RLS) estimator and a GPC predictor is discussed here.

Parallel implementation of a learning algorithm for multilayer perceptrons using transputers

In this paper the parallelization of a new learning algorithm for multilayer perceptrons, specifically targeted for nonlinear function approximation purposes, is discussed. Each major step of the algorithm is parallelized, a special emphasis being put in the most computationally intensive task, a least-squares solution of linear systems of equations.

Preconditioning 2D Integer Data for Fast Convex Hull Computations

In order to accelerate computing the convex hull on a set of n points, a heuristic procedure is often applied to reduce the number of points to a set of s points, s ≤ n, which also contains the same hull. We present an algorithm to precondition 2D data with integer coordinates bounded by a box of size p × q before building a 2D convex hull, with three distinct advantages. First, we prove that under the condition min(p, q) ≤ n the algorithm executes in time within O(n); second, no explicit sorting of data is required; and third, the reduced set of s points forms a simple polygonal chain and thus can be directly pipelined into an O(n) time convex hull algorithm. This paper empirically evaluates and quantifies the speed up gained by preconditioning a set of points by a method based on the proposed algorithm before using common convex hull algorithms to build the final hull. A speedup factor of at least four is consistently found from experiments on various datasets when the condition min(p, q) ≤ n holds; the smaller the ratio min(p, q)/n is in the dataset, the greater the speedup factor achieved.

A Study of Ordered Gene Problems Featuring DNA Error Correction and DNA Fragment Assembly with a Variety of Heuristics, Genetic Algorithm Variations, and Dynamic Representations

Ordered gene problems are a very common classification of optimization problems. Because of their popularity countless algorithms have been developed in an attempt to find high quality solutions to the problems. It is also common to see many different types of problems reduced to ordered gene style problems as there are many popular heuristics and metaheuristics for them due to their popularity. Multiple ordered gene problems are studied, namely, the travelling salesman problem, bin packing problem, and graph colouring problem. In addition, two bioinformatics problems not traditionally seen as ordered gene problems are studied: DNA error correction and DNA fragment assembly. These problems are studied with multiple variations and combinations of heuristics and metaheuristics with two distinct types or representations. The majority of the algorithms are built around the Recentering- Restarting Genetic Algorithm. The algorithm variations were successful on all problems studied, and particularly for the two bioinformatics problems. For DNA Error Correction multiple cases were found with 100% of the codes being corrected. The algorithm variations were also able to beat all other state-of-the-art DNA Fragment Assemblers on 13 out of 16 benchmark problem instances.

Incorporating complex cells into neural networks for pattern classification

Dans le domaine des neurosciences computationnelles, l'hypothèse a été émise que le système visuel, depuis la rétine et jusqu'au cortex visuel primaire au moins, ajuste continuellement un modèle probabiliste avec des variables latentes, à son flux de perceptions. Ni le modèle exact, ni la méthode exacte utilisée pour l'ajustement ne sont connus, mais les algorithmes existants qui permettent l'ajustement de tels modèles ont besoin de faire une estimation conditionnelle des variables latentes. Cela nous peut nous aider à comprendre pourquoi le système visuel pourrait ajuster un tel modèle; si le modèle est approprié, ces estimé conditionnels peuvent aussi former une excellente représentation, qui permettent d'analyser le contenu sémantique des images perçues. Le travail présenté ici utilise la performance en classification d'images (discrimination entre des types d'objets communs) comme base pour comparer des modèles du système visuel, et des algorithmes pour ajuster ces modèles (vus comme des densités de probabilité) à des images. Cette thèse (a) montre que des modèles basés sur les cellules complexes de l'aire visuelle V1 généralisent mieux à partir d'exemples d'entraînement étiquetés que les réseaux de neurones conventionnels, dont les unités cachées sont plus semblables aux cellules simples de V1; (b) présente une nouvelle interprétation des modèles du système visuels basés sur des cellules complexes, comme distributions de probabilités, ainsi que de nouveaux algorithmes pour les ajuster à des données; et (c) montre que ces modèles forment des représentations qui sont meilleures pour la classification d'images, après avoir été entraînés comme des modèles de probabilités. Deux innovations techniques additionnelles, qui ont rendu ce travail possible, sont également décrites : un algorithme de recherche aléatoire pour sélectionner des hyper-paramètres, et un compilateur pour des expressions mathématiques matricielles, qui peut optimiser ces expressions pour processeur central (CPU) et graphique (GPU).

Méthode numérique d'estimation du mouvement des masses molles

L’analyse biomécanique du mouvement humain en utilisant des systèmes optoélectroniques et des marqueurs cutanés considère les segments du corps comme des corps rigides. Cependant, le mouvement des tissus mous par rapport à l'os, c’est à dire les muscles et le tissu adipeux, provoque le déplacement des marqueurs. Ce déplacement est le fait de deux composantes, une composante propre correspondant au mouvement aléatoire de chaque marqueur et une composante à l’unisson provoquant le déplacement commun des marqueurs cutanés lié au mouvement des masses sous-jacentes. Si nombre d’études visent à minimiser ces déplacements, des simulations ont montré que le mouvement des masses molles réduit la dynamique articulaire. Cette observation est faite uniquement par la simulation, car il n'existe pas de méthodes capables de dissocier la cinématique des masses molles de celle de l’os. L’objectif principal de cette thèse consiste à développer une méthode numérique capable de distinguer ces deux cinématiques. Le premier objectif était d'évaluer une méthode d'optimisation locale pour estimer le mouvement des masses molles par rapport à l’humérus obtenu avec une tige intra-corticale vissée chez trois sujets. Les résultats montrent que l'optimisation locale sous-estime de 50% le déplacement des marqueurs et qu’elle conduit à un classement de marqueurs différents en fonction de leur déplacement. La limite de cette méthode vient du fait qu'elle ne tient pas compte de l’ensemble des composantes du mouvement des tissus mous, notamment la composante en unisson. Le second objectif était de développer une méthode numérique qui considère toutes les composantes du mouvement des tissus mous. Plus précisément, cette méthode devait fournir une cinématique similaire et une plus grande estimation du déplacement des marqueurs par rapport aux méthodes classiques et dissocier ces composantes. Le membre inférieur est modélisé avec une chaine cinématique de 10 degrés de liberté reconstruite par optimisation globale en utilisant seulement les marqueurs placés sur le pelvis et la face médiale du tibia. L’estimation de la cinématique sans considérer les marqueurs placés sur la cuisse et le mollet permet d'éviter l’influence de leur déplacement sur la reconstruction du modèle cinématique. Cette méthode testée sur 13 sujets lors de sauts a obtenu jusqu’à 2,1 fois plus de déplacement des marqueurs en fonction de la méthode considérée en assurant des cinématiques similaires. Une approche vectorielle a montré que le déplacement des marqueurs est surtout dû à la composante à l’unisson. Une approche matricielle associant l’optimisation locale à la chaine cinématique a montré que les masses molles se déplacent principalement autour de l'axe longitudinal et le long de l'axe antéro-postérieur de l'os. L'originalité de cette thèse est de dissocier numériquement la cinématique os de celle des masses molles et les composantes de ce mouvement. Les méthodes développées dans cette thèse augmentent les connaissances sur le mouvement des masses molles et permettent d’envisager l’étude de leur effet sur la dynamique articulaire.

An algorithm for identifying agent-k-linked allocations in economies with indivisibilities

We consider envy-free (and budget-balanced) rules that are least manipulable with respect to agents counting or with respect to utility gains. Recently it has been shown that for any profile of quasi-linear preferences, the outcome of any such least manipulable envy-free rule can be obtained via agent-k-linked allocations. This note provides an algorithm for identifying agent-k-linked allocations.

Evolving Distributed Algorithms with Genetic Programming

Distributed systems are one of the most vital components of the economy. The most prominent example is probably the internet, a constituent element of our knowledge society. During the recent years, the number of novel network types has steadily increased. Amongst others, sensor networks, distributed systems composed of tiny computational devices with scarce resources, have emerged. The further development and heterogeneous connection of such systems imposes new requirements on the software development process. Mobile and wireless networks, for instance, have to organize themselves autonomously and must be able to react to changes in the environment and to failing nodes alike. Researching new approaches for the design of distributed algorithms may lead to methods with which these requirements can be met efficiently. In this thesis, one such method is developed, tested, and discussed in respect of its practical utility. Our new design approach for distributed algorithms is based on Genetic Programming, a member of the family of evolutionary algorithms. Evolutionary algorithms are metaheuristic optimization methods which copy principles from natural evolution. They use a population of solution candidates which they try to refine step by step in order to attain optimal values for predefined objective functions. The synthesis of an algorithm with our approach starts with an analysis step in which the wanted global behavior of the distributed system is specified. From this specification, objective functions are derived which steer a Genetic Programming process where the solution candidates are distributed programs. The objective functions rate how close these programs approximate the goal behavior in multiple randomized network simulations. The evolutionary process step by step selects the most promising solution candidates and modifies and combines them with mutation and crossover operators. This way, a description of the global behavior of a distributed system is translated automatically to programs which, if executed locally on the nodes of the system, exhibit this behavior. In our work, we test six different ways for representing distributed programs, comprising adaptations and extensions of well-known Genetic Programming methods (SGP, eSGP, and LGP), one bio-inspired approach (Fraglets), and two new program representations called Rule-based Genetic Programming (RBGP, eRBGP) designed by us. We breed programs in these representations for three well-known example problems in distributed systems: election algorithms, the distributed mutual exclusion at a critical section, and the distributed computation of the greatest common divisor of a set of numbers. Synthesizing distributed programs the evolutionary way does not necessarily lead to the envisaged results. In a detailed analysis, we discuss the problematic features which make this form of Genetic Programming particularly hard. The two Rule-based Genetic Programming approaches have been developed especially in order to mitigate these difficulties. In our experiments, at least one of them (eRBGP) turned out to be a very efficient approach and in most cases, was superior to the other representations.

Approximate Gauss-Newton methods for nonlinear least squares problems

The Gauss–Newton algorithm is an iterative method regularly used for solving nonlinear least squares problems. It is particularly well suited to the treatment of very large scale variational data assimilation problems that arise in atmosphere and ocean forecasting. The procedure consists of a sequence of linear least squares approximations to the nonlinear problem, each of which is solved by an “inner” direct or iterative process. In comparison with Newton’s method and its variants, the algorithm is attractive because it does not require the evaluation of second-order derivatives in the Hessian of the objective function. In practice the exact Gauss–Newton method is too expensive to apply operationally in meteorological forecasting, and various approximations are made in order to reduce computational costs and to solve the problems in real time. Here we investigate the effects on the convergence of the Gauss–Newton method of two types of approximation used commonly in data assimilation. First, we examine “truncated” Gauss–Newton methods where the inner linear least squares problem is not solved exactly, and second, we examine “perturbed” Gauss–Newton methods where the true linearized inner problem is approximated by a simplified, or perturbed, linear least squares problem. We give conditions ensuring that the truncated and perturbed Gauss–Newton methods converge and also derive rates of convergence for the iterations. The results are illustrated by a simple numerical example. A practical application to the problem of data assimilation in a typical meteorological system is presented.

A geometric algorithm for an artificial life form based on anachronistic methods

Modern methods of spawning new technological motifs are not appropriate when it is desired to realize artificial life as an actual real world entity unto itself (Pattee 1995; Brooks 2006; Chalmers 1995). Many fundamental aspects of such a machine are absent in common methods, which generally lack methodologies of construction. In this paper we mix classical and modern studies in order to attempt to realize an artificial life form from first principles. A model of an algorithm is introduced, its methodology of construction is presented, and the fundamental source from which it sprang is discussed.

Dynamic load balancing for the distributed mining of molecular structures

In molecular biology, it is often desirable to find common properties in large numbers of drug candidates. One family of methods stems from the data mining community, where algorithms to find frequent graphs have received increasing attention over the past years. However, the computational complexity of the underlying problem and the large amount of data to be explored essentially render sequential algorithms useless. In this paper, we present a distributed approach to the frequent subgraph mining problem to discover interesting patterns in molecular compounds. This problem is characterized by a highly irregular search tree, whereby no reliable workload prediction is available. We describe the three main aspects of the proposed distributed algorithm, namely, a dynamic partitioning of the search space, a distribution process based on a peer-to-peer communication framework, and a novel receiverinitiated load balancing algorithm. The effectiveness of the distributed method has been evaluated on the well-known National Cancer Institute’s HIV-screening data set, where we were able to show close-to linear speedup in a network of workstations. The proposed approach also allows for dynamic resource aggregation in a non dedicated computational environment. These features make it suitable for large-scale, multi-domain, heterogeneous environments, such as computational grids.

High performance subgraph mining in molecular compounds

Structured data represented in the form of graphs arises in several fields of the science and the growing amount of available data makes distributed graph mining techniques particularly relevant. In this paper, we present a distributed approach to the frequent subgraph mining problem to discover interesting patterns in molecular compounds. The problem is characterized by a highly irregular search tree, whereby no reliable workload prediction is available. We describe the three main aspects of the proposed distributed algorithm, namely a dynamic partitioning of the search space, a distribution process based on a peer-to-peer communication framework, and a novel receiver-initiated, load balancing algorithm. The effectiveness of the distributed method has been evaluated on the well-known National Cancer Institute’s HIV-screening dataset, where the approach attains close-to linear speedup in a network of workstations.

The method of predicate least squares refinement

Parameters to be determined in a least squares refinement calculation to fit a set of observed data may sometimes usefully be `predicated' to values obtained from some independent source, such as a theoretical calculation. An algorithm for achieving this in a least squares refinement calculation is described, which leaves the operator in full control of the weight that he may wish to attach to the predicate values of the parameters.

Phylogeographic analysis of the chloroplast DNA variation in wild common bean (Phaseolus vulgaris L.) in the Americas

The wild common bean (Phaseolus vulgaris) is widely but discontinuously distributed from northern Mexico to northern Argentina on both sides of the Isthmus of Panama. Little is known on how the species has reached its current disjunct distribution. In this research, chloroplast DNA polymorphisms in seven non-coding regions were used to study the history of migration of wild P. vulgaris between Mesoamerica and South America. A penalized likelihood analysis was applied to previously published Leguminosae ITS data to estimate divergence times between P. vulgaris and its sister taxa from Mesoamerica, and divergence times of populations within P. vulgaris. Fourteen chloroplast haplotypes were identified by PCR-RFLP and their geographical associations were studied by means of a Nested Clade Analysis and Mantel Tests. The results suggest that the haplotypes are not randomly distributed but occupy discrete parts of the geographic range of the species. The current distribution of haplotypes may be explained by isolation by distance and by at least two migration events between Mesoamerica and South America: one from Mesoamerica to South America and another one from northern South America to Mesoamerica. Age estimates place the divergence of P. vulgaris from its sister taxa from Mesoamerica at or before 1.3 Ma, and divergence of populations from Ecuador-northern Peru at or before 0.6 Ma. As these ages are taken as minimum divergence times, the influence of past events, such as the closure of the Isthmus of Panama and the final uplift of the Andes, on the migration history and population structure of this species cannot be disregarded.

The visibility of agricultural subsidies and market illusions in the Common Agricultural Policy: Some evidence from farmers' views in Germany, Portugal and the United Kingdom

This article demonstrates that the design and nature of agricultural support schemes has an influence on farmers' perception of their level of dependence on agricultural support. While direct aid payments inform farmers about the extent to which they are subsidised, indirect support mechanisms veil the level of subsidisation, and therefore they are not fully aware of the extent to which they are supported. To test this hypothesis, we applied data from a survey of 4,500 farmers in three countries: the United Kingdom, Germany and Portugal. It is demonstrated that indirect support, such as that provided through artificially high consumer prices, gives an illusion of free and competitive markets among farmers. This 'visibility' hypothesis is evaluated against an alternative hypothesis that assumes farmers have complete, or at least a fairly comprehensive level of, information on agricultural support schemes. Our findings show that this alternative hypothesis can be ruled out.