Design and characterization of large area position sensitive radiation detectors

Planar, large area, position sensitive silicon detectors are widely utilized in high energy physics research and in medical, computed tomography (CT). This thesis describes author's research work relating to development of such detector components. The key motivation and objective for the research work has been the development of novel, position sensitive detectors improving the performance of the instruments they are intended for. Silicon strip detectors are the key components of barrel-shaped tracking instruments which are typically the innermost structures of high energy physics experimental stations. Particle colliders such as the former LEP collider or present LHC produce particle collisions and the silicon strip detector based trackers locate the trajectories of particles emanating from such collisions. Medical CT has become a regular part of everyday medical care in all developed countries. CT scanning enables x-ray imaging of all parts of the human body with an outstanding structural resolution and contrast. Brain, chest and abdomen slice images with a resolution of 0.5 mm are possible and latest CT machines are able to image whole human heart between heart beats. The two application areas are presented shortly and the radiation detection properties of planar silicon detectors are discussed. Fabrication methods and preamplifier electronics of the planar detectors are presented. Designs of the developed, large area silicon detectors are presented and measurement results of the key operating parameters are discussed. Static and dynamic performance of the developed silicon strip detectors are shown to be very satisfactory for experimental physics applications. Results relating to the developed, novel CT detector chips are found to be very promising for further development and all key performance goals are met.

Optimization and Measuring Techniques for Collect-and-Place Machines in Printed Circuit Board Industry

This thesis considers optimization problems arising in printed circuit board assembly. Especially, the case in which the electronic components of a single circuit board are placed using a single placement machine is studied. Although there is a large number of different placement machines, the use of collect-and-place -type gantry machines is discussed because of their flexibility and increasing popularity in the industry. Instead of solving the entire control optimization problem of a collect-andplace machine with a single application, the problem is divided into multiple subproblems because of its hard combinatorial nature. This dividing technique is called hierarchical decomposition. All the subproblems of the one PCB - one machine -context are described, classified and reviewed. The derived subproblems are then either solved with exact methods or new heuristic algorithms are developed and applied. The exact methods include, for example, a greedy algorithm and a solution based on dynamic programming. Some of the proposed heuristics contain constructive parts while others utilize local search or are based on frequency calculations. For the heuristics, it is made sure with comprehensive experimental tests that they are applicable and feasible. A number of quality functions will be proposed for evaluation and applied to the subproblems. In the experimental tests, artificially generated data from Markov-models and data from real-world PCB production are used. The thesis consists of an introduction and of five publications where the developed and used solution methods are described in their full detail. For all the problems stated in this thesis, the methods proposed are efficient enough to be used in the PCB assembly production in practice and are readily applicable in the PCB manufacturing industry.

Improving sampling, optimization and feature extraction in Boltzmann machines

L’apprentissage supervisé de réseaux hiérarchiques à grande échelle connaît présentement un succès fulgurant. Malgré cette effervescence, l’apprentissage non-supervisé représente toujours, selon plusieurs chercheurs, un élément clé de l’Intelligence Artificielle, où les agents doivent apprendre à partir d’un nombre potentiellement limité de données. Cette thèse s’inscrit dans cette pensée et aborde divers sujets de recherche liés au problème d’estimation de densité par l’entremise des machines de Boltzmann (BM), modèles graphiques probabilistes au coeur de l’apprentissage profond. Nos contributions touchent les domaines de l’échantillonnage, l’estimation de fonctions de partition, l’optimisation ainsi que l’apprentissage de représentations invariantes. Cette thèse débute par l’exposition d’un nouvel algorithme d'échantillonnage adaptatif, qui ajuste (de fa ̧con automatique) la température des chaînes de Markov sous simulation, afin de maintenir une vitesse de convergence élevée tout au long de l’apprentissage. Lorsqu’utilisé dans le contexte de l’apprentissage par maximum de vraisemblance stochastique (SML), notre algorithme engendre une robustesse accrue face à la sélection du taux d’apprentissage, ainsi qu’une meilleure vitesse de convergence. Nos résultats sont présent ́es dans le domaine des BMs, mais la méthode est générale et applicable à l’apprentissage de tout modèle probabiliste exploitant l’échantillonnage par chaînes de Markov. Tandis que le gradient du maximum de vraisemblance peut-être approximé par échantillonnage, l’évaluation de la log-vraisemblance nécessite un estimé de la fonction de partition. Contrairement aux approches traditionnelles qui considèrent un modèle donné comme une boîte noire, nous proposons plutôt d’exploiter la dynamique de l’apprentissage en estimant les changements successifs de log-partition encourus à chaque mise à jour des paramètres. Le problème d’estimation est reformulé comme un problème d’inférence similaire au filtre de Kalman, mais sur un graphe bi-dimensionnel, où les dimensions correspondent aux axes du temps et au paramètre de température. Sur le thème de l’optimisation, nous présentons également un algorithme permettant d’appliquer, de manière efficace, le gradient naturel à des machines de Boltzmann comportant des milliers d’unités. Jusqu’à présent, son adoption était limitée par son haut coût computationel ainsi que sa demande en mémoire. Notre algorithme, Metric-Free Natural Gradient (MFNG), permet d’éviter le calcul explicite de la matrice d’information de Fisher (et son inverse) en exploitant un solveur linéaire combiné à un produit matrice-vecteur efficace. L’algorithme est prometteur: en terme du nombre d’évaluations de fonctions, MFNG converge plus rapidement que SML. Son implémentation demeure malheureusement inefficace en temps de calcul. Ces travaux explorent également les mécanismes sous-jacents à l’apprentissage de représentations invariantes. À cette fin, nous utilisons la famille de machines de Boltzmann restreintes “spike & slab” (ssRBM), que nous modifions afin de pouvoir modéliser des distributions binaires et parcimonieuses. Les variables latentes binaires de la ssRBM peuvent être rendues invariantes à un sous-espace vectoriel, en associant à chacune d’elles, un vecteur de variables latentes continues (dénommées “slabs”). Ceci se traduit par une invariance accrue au niveau de la représentation et un meilleur taux de classification lorsque peu de données étiquetées sont disponibles. Nous terminons cette thèse sur un sujet ambitieux: l’apprentissage de représentations pouvant séparer les facteurs de variations présents dans le signal d’entrée. Nous proposons une solution à base de ssRBM bilinéaire (avec deux groupes de facteurs latents) et formulons le problème comme l’un de “pooling” dans des sous-espaces vectoriels complémentaires.

Notes on PCA, Regularization, Sparsity and Support Vector Machines

We derive a new representation for a function as a linear combination of local correlation kernels at optimal sparse locations and discuss its relation to PCA, regularization, sparsity principles and Support Vector Machines. We first review previous results for the approximation of a function from discrete data (Girosi, 1998) in the context of Vapnik"s feature space and dual representation (Vapnik, 1995). We apply them to show 1) that a standard regularization functional with a stabilizer defined in terms of the correlation function induces a regression function in the span of the feature space of classical Principal Components and 2) that there exist a dual representations of the regression function in terms of a regularization network with a kernel equal to a generalized correlation function. We then describe the main observation of the paper: the dual representation in terms of the correlation function can be sparsified using the Support Vector Machines (Vapnik, 1982) technique and this operation is equivalent to sparsify a large dictionary of basis functions adapted to the task, using a variation of Basis Pursuit De-Noising (Chen, Donoho and Saunders, 1995; see also related work by Donahue and Geiger, 1994; Olshausen and Field, 1995; Lewicki and Sejnowski, 1998). In addition to extending the close relations between regularization, Support Vector Machines and sparsity, our work also illuminates and formalizes the LFA concept of Penev and Atick (1996). We discuss the relation between our results, which are about regression, and the different problem of pattern classification.

Support Vector Machines: Training and Applications

The Support Vector Machine (SVM) is a new and very promising classification technique developed by Vapnik and his group at AT&T Bell Labs. This new learning algorithm can be seen as an alternative training technique for Polynomial, Radial Basis Function and Multi-Layer Perceptron classifiers. An interesting property of this approach is that it is an approximate implementation of the Structural Risk Minimization (SRM) induction principle. The derivation of Support Vector Machines, its relationship with SRM, and its geometrical insight, are discussed in this paper. Training a SVM is equivalent to solve a quadratic programming problem with linear and box constraints in a number of variables equal to the number of data points. When the number of data points exceeds few thousands the problem is very challenging, because the quadratic form is completely dense, so the memory needed to store the problem grows with the square of the number of data points. Therefore, training problems arising in some real applications with large data sets are impossible to load into memory, and cannot be solved using standard non-linear constrained optimization algorithms. We present a decomposition algorithm that can be used to train SVM's over large data sets. The main idea behind the decomposition is the iterative solution of sub-problems and the evaluation of, and also establish the stopping criteria for the algorithm. We present previous approaches, as well as results and important details of our implementation of the algorithm using a second-order variant of the Reduced Gradient Method as the solver of the sub-problems. As an application of SVM's, we present preliminary results we obtained applying SVM to the problem of detecting frontal human faces in real images.

Hierarchical Prediction Machines and Big Data Analytics

An emerging consensus in cognitive science views the biological brain as a hierarchically-organized predictive processing system. This is a system in which higher-order regions are continuously attempting to predict the activity of lower-order regions at a variety of (increasingly abstract) spatial and temporal scales. The brain is thus revealed as a hierarchical prediction machine that is constantly engaged in the effort to predict the flow of information originating from the sensory surfaces. Such a view seems to afford a great deal of explanatory leverage when it comes to a broad swathe of seemingly disparate psychological phenomena (e.g., learning, memory, perception, action, emotion, planning, reason, imagination, and conscious experience). In the most positive case, the predictive processing story seems to provide our first glimpse at what a unified (computationally-tractable and neurobiological plausible) account of human psychology might look like. This obviously marks out one reason why such models should be the focus of current empirical and theoretical attention. Another reason, however, is rooted in the potential of such models to advance the current state-of-the-art in machine intelligence and machine learning. Interestingly, the vision of the brain as a hierarchical prediction machine is one that establishes contact with work that goes under the heading of 'deep learning'. Deep learning systems thus often attempt to make use of predictive processing schemes and (increasingly abstract) generative models as a means of supporting the analysis of large data sets. But are such computational systems sufficient (by themselves) to provide a route to general human-level analytic capabilities? I will argue that they are not and that closer attention to a broader range of forces and factors (many of which are not confined to the neural realm) may be required to understand what it is that gives human cognition its distinctive (and largely unique) flavour. The vision that emerges is one of 'homomimetic deep learning systems', systems that situate a hierarchically-organized predictive processing core within a larger nexus of developmental, behavioural, symbolic, technological and social influences. Relative to that vision, I suggest that we should see the Web as a form of 'cognitive ecology', one that is as much involved with the transformation of machine intelligence as it is with the progressive reshaping of our own cognitive capabilities.

Purpose built teaching equipment for a laboratory based course in three phase machines and drives

The hazards associated with high voltage three phase inverters and the rotating shafts of large electrical machines have resulted in most of the engineering courses covering these topics to be predominantly theoretical. This paper describes a set of purpose built, low voltage and low cost teaching equipment which allows the "hands on" instruction of three phase inverters and rotating machines. By using low voltages, the student can experiment freely with the motors and inverter and can access all of the current and voltage waveforms, which until now could only be studied in text books or observed as part of laboratory demonstrations. Both the motor and the inverter designs are optimized for teaching purposes cost around $25 and can be made with minimal effort.

Three-phase machines and drives - equipment for a laboratory-based course

The hazards associated with high-voltage three-phase inverters and high-powered large electrical machines have resulted in most of the engineering courses covering three-phase machines and drives theoretically. This paper describes a set of purpose-built, low-voltage, and low-cost teaching equipment that allows the hands-on instruction of three-phase inverters and rotating machines. The motivation for moving towards a system running at low voltages is that the students can safely experiment freely with the motors and inverter. The students can also access all of the current and voltage waveforms, which until now could only be studied in textbooks or observed as part of laboratory demonstrations. Both the motor and the inverter designs are for teaching purposes and require minimal effort and cost

Purpose built teaching equipment for a laboratory based course in three phase machines and drives

The hazards associated with high voltage three phase inverters ond the rotating sha@s of large electrical machines have resulted in most of the engineering courses covering these topics to be predominantly theoretical. This paper describes a set of purpose built, low voltage and low cost teaching equipment which allows the “hands on I’ instruction of three phase inverters and rotating machines. By using low voltages, the student can experiment freely with the motors and inverter and can access all of the current and voltage waveforms, which until now could only be studied in text books or observed as part of laboratory demonstrations. Both the motor and the inverter designs are optimized for teaching purposes, cost around $25 and can be made with minimal effort.

Three-phase machines and drives - Equipment for a laboratory-based course

The hazards associated with high-voltage three-phase inverters and high-powered large electrical machines have resulted in most of the engineering courses covering three-phase machines and drives theoretically. This paper describes a set of purpose-built, low-voltage, and low-cost teaching equipment that allows the hands-on instruction of three-phase inverters and rotating machines. The motivation for moving towards a system running at low voltages is that the students can safely experiment freely with the motors and inverter. The students can also access all of the current and voltage waveforms, which until now could only be studied in textbooks or observed as part of laboratory demonstrations. Both the motor and the inverter designs are for teaching purposes and require minimal effort and cost.

Landscape state machines: Tools for evolutionary algorithm performance analyses and landscape/algorithm mapping

Many evolutionary algorithm applications involve either fitness functions with high time complexity or large dimensionality (hence very many fitness evaluations will typically be needed) or both. In such circumstances, there is a dire need to tune various features of the algorithm well so that performance and time savings are optimized. However, these are precisely the circumstances in which prior tuning is very costly in time and resources. There is hence a need for methods which enable fast prior tuning in such cases. We describe a candidate technique for this purpose, in which we model a landscape as a finite state machine, inferred from preliminary sampling runs. In prior algorithm-tuning trials, we can replace the 'real' landscape with the model, enabling extremely fast tuning, saving far more time than was required to infer the model. Preliminary results indicate much promise, though much work needs to be done to establish various aspects of the conditions under which it can be most beneficially used. A main limitation of the method as described here is a restriction to mutation-only algorithms, but there are various ways to address this and other limitations.

Design and analysis of a haptic device design for large and fast movements

Haptic devices tend to be kept small as it is easier to achieve a large change of stiffness with a low associated apparent mass. If large movements are required there is a usually a reduction in the quality of the haptic sensations which can be displayed. The typical measure of haptic device performance is impedance-width (z-width) but this does not account for actuator saturation, usable workspace or the ability to do rapid movements. This paper presents the analysis and evaluation of a haptic device design, utilizing a variant of redundant kinematics, sometimes referred to as a macro-micro configuration, intended to allow large and fast movements without loss of impedance-width. A brief mathematical analysis of the design constraints is given and a prototype system is described where the effects of different elements of the control scheme can be examined to better understand the potential benefits and trade-offs in the design. Finally, the performance of the system is evaluated using a Fitts’ Law test and found to compare favourably with similar evaluations of smaller workspace devices.

Checking Completeness of Tests for Finite State Machines

In testing from a Finite State Machine (FSM), the generation of test suites which guarantee full fault detection, known as complete test suites, has been a long-standing research topic. In this paper, we present conditions that are sufficient for a test suite to be complete. We demonstrate that the existing conditions are special cases of the proposed ones. An algorithm that checks whether a given test suite is complete is given. The experimental results show that the algorithm can be used for relatively large FSMs and test suites.

Efficient supervised optimum-path forest classification for large datasets

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Fault detection in rotating machines by vibration signal processing techniques

Machines with moving parts give rise to vibrations and consequently noise. The setting up and the status of each machine yield to a peculiar vibration signature. Therefore, a change in the vibration signature, due to a change in the machine state, can be used to detect incipient defects before they become critical. This is the goal of condition monitoring, in which the informations obtained from a machine signature are used in order to detect faults at an early stage. There are a large number of signal processing techniques that can be used in order to extract interesting information from a measured vibration signal. This study seeks to detect rotating machine defects using a range of techniques including synchronous time averaging, Hilbert transform-based demodulation, continuous wavelet transform, Wigner-Ville distribution and spectral correlation density function. The detection and the diagnostic capability of these techniques are discussed and compared on the basis of experimental results concerning gear tooth faults, i.e. fatigue crack at the tooth root and tooth spalls of different sizes, as well as assembly faults in diesel engine. Moreover, the sensitivity to fault severity is assessed by the application of these signal processing techniques to gear tooth faults of different sizes.