Algorithm engineering for expression dag based number types

Magdeburg, Univ., Fak. für Informatik, Diss., 2015

Efficient query processing in co-processor-accelerated databases

...Diese Dissertation zeigt, wie wir Datenbankmanagementsysteme bauen können, die heterogene Prozessoren effizient und zuverlässig zur Beschleunigung der Anfrageverarbeitung nutzen können. Daher untersuchen wir typische Entwurfsentscheidungen von coprozessorbeschleunigten Datenbankmanagementsystemen und leiten darauf aufbauend eine generische Architektur für solche Systeme ab. Unsere Untersuchungen zeigen, dass eines der wichtigsten Probleme für solche Datenbankmanagementsysteme die Entscheidung ist, welche Operatoren einer Anfrage auf welchem Prozessor ausgeführt werden sollen...

A note on parallelizing the parameterized expectations algorithm

The parameterized expectations algorithm (PEA) involves a long simulation and a nonlinear least squares (NLS) fit, both embedded in a loop. Both steps are natural candidates for parallelization. This note shows that parallelization can lead to important speedups for the PEA. I provide example code for a simple model that can serve as a template for parallelization of more interesting models, as well as a download link for an image of a bootable CD that allows creation of a cluster and execution of the example code in minutes, with no need to install any software.

Efficient bidding with externalities

We implement a family of efficient proposals to share benefits generated in environments with externalities. These proposals extend the Shapley value to games with externalities and are parametrized through the method by which the externalities are averaged. We construct two slightly different mechanisms: one for environments with negative externalities and the other for positive externalities. We show that the subgame perfect equilibrium outcomes of these mechanisms coincide with the sharing proposals.

Fair and efficient student placement with couples

We study situations of allocating positions or jobs to students or workers based on priorities. An example is the assignment of medical students to hospital residencies on the basis of one or several entrance exams. For markets without couples, e.g., for ``undergraduate student placement,'' acyclicity is a necessary and sufficient condition for the existence of a fair and efficient placement mechanism (Ergin, 2002). We show that in the presence of couples, which introduces complementarities into the students' preferences, acyclicity is still necessary, but not sufficient (Theorem 4.1). A second necessary condition (Theorem 4.2) is ``priority-togetherness'' of couples. A priority structure that satisfies both necessary conditions is called pt-acyclic. For student placement problems where all quotas are equal to one we characterize pt-acyclicity (Lemma 5.1) and show that it is a sufficient condition for the existence of a fair and efficient placement mechanism (Theorem 5.1). If in addition to pt-acyclicity we require ``reallocation-'' and ``vacancy-fairness'' for couples, the so-called dictator-bidictator placement mechanism is the unique fair and efficient placement mechanism (Theorem 5.2). Finally, for general student placement problems, we show that pt-acyclicity may not be sufficient for the existence of a fair and efficient placement mechanism (Examples 5.4, 5.5, and 5.6). We identify a sufficient condition such that the so-called sequential placement mechanism produces a fair and efficient allocation (Theorem 5.3).

Sharing the surplus: a just and efficient proposal for environments with externalities

Economic activities, both on the macro and micro level, often entail wide-spread externalities. This in turn leads to disputes regarding the compensation levels to the various parties affected. We propose a general, yet simple, method of deciding upon the distribution of the gains (costs) of cooperation in the presence of externalities. This method is shown to be the unique one satisfying several desirable properties. Furthermore, we illustrate the use of this method to resolve the sharing of benefits generated by international climate control agreements.

Forming efficient networks

In this paper, we suggest a simple sequential mechanism whose subgame perfect equilibria give rise to efficient networks. Moreover, the payoffs received by the agents coincide with their Shapley value in an appropriately defined cooperative game.

Bidding for the surplus: realizing efficient outcomes in general economic environments

In this paper, we consider two classes of economic environments. In the first type, agents are faced with the task of providing local public goods that will benefit some or all of them. In the second type, economic activity takes place via formation of links. Agents need both to both form a network and decide how to share the output generated. For both scenarios, we suggest a bidding mechanism whereby agents bid for the right to decide upon the organization of the economic activity. The subgame perfect equilibria of this game generate efficient outcomes.

Efficient and stable collective choices under gregarious preferences

We consider collective choice problems where a set of agents have to choose an alternative from a finite set and agents may or may not become users of the chosen alternative. An allocation is a pair given by the chosen alternative and the set of its users. Agents have gregarious preferences over allocations: given an allocation, they prefer that the set of users becomes larger. We require that the final allocation be efficient and stable (no agent can be forced to be a user and no agent who wants to be a user can be excluded). We propose a two-stage sequential mechanism whose unique subgame perfect equilibrium outcome is an efficient and stable allocation which also satisfies a maximal participation property.

Multi-scale support vector algorithms for hot spot detection and modelling

The algorithmic approach to data modelling has developed rapidly these last years, in particular methods based on data mining and machine learning have been used in a growing number of applications. These methods follow a data-driven methodology, aiming at providing the best possible generalization and predictive abilities instead of concentrating on the properties of the data model. One of the most successful groups of such methods is known as Support Vector algorithms. Following the fruitful developments in applying Support Vector algorithms to spatial data, this paper introduces a new extension of the traditional support vector regression (SVR) algorithm. This extension allows for the simultaneous modelling of environmental data at several spatial scales. The joint influence of environmental processes presenting different patterns at different scales is here learned automatically from data, providing the optimum mixture of short and large-scale models. The method is adaptive to the spatial scale of the data. With this advantage, it can provide efficient means to model local anomalies that may typically arise in situations at an early phase of an environmental emergency. However, the proposed approach still requires some prior knowledge on the possible existence of such short-scale patterns. This is a possible limitation of the method for its implementation in early warning systems. The purpose of this paper is to present the multi-scale SVR model and to illustrate its use with an application to the mapping of Cs137 activity given the measurements taken in the region of Briansk following the Chernobyl accident.

Efficient knowledge retrieval to calibrate input variables in forest fire prediction

Forest fires are a serious threat to humans and nature from an ecological, social and economic point of view. Predicting their behaviour by simulation still delivers unreliable results and remains a challenging task. Latest approaches try to calibrate input variables, often tainted with imprecision, using optimisation techniques like Genetic Algorithms. To converge faster towards fitter solutions, the GA is guided with knowledge obtained from historical or synthetical fires. We developed a robust and efficient knowledge storage and retrieval method. Nearest neighbour search is applied to find the fire configuration from knowledge base most similar to the current configuration. Therefore, a distance measure was elaborated and implemented in several ways. Experiments show the performance of the different implementations regarding occupied storage and retrieval time with overly satisfactory results.

Assessing the assignation of public subsidies: Do the experts choose the most efficient R&D projects?

The implementation of public programs to support business R&D projects requires the establishment of a selection process. This selection process faces various difficulties, which include the measurement of the impact of the R&D projects as well as selection process optimization among projects with multiple, and sometimes incomparable, performance indicators. To this end, public agencies generally use the peer review method, which, while presenting some advantages, also demonstrates significant drawbacks. Private firms, on the other hand, tend toward more quantitative methods, such as Data Envelopment Analysis (DEA), in their pursuit of R&D investment optimization. In this paper, the performance of a public agency peer review method of project selection is compared with an alternative DEA method.

Are two-part tariffs efficient when consumers plan ahead?: An empirical study

During the last two decades there has been an increase in using dynamic tariffs for billing household electricity consumption. This has questioned the suitability of traditional pricing schemes, such as two-part tariffs, since they contribute to create marked peak and offpeak demands. The aim of this paper is to assess if two-part tariffs are an efficient pricing scheme using Spanish household electricity microdata. An ordered probit model with instrumental variables on the determinants of power level choice and non-paramentric spline regressions on the electricity price distribution will allow us to distinguish between the tariff structure choice and the simultaneous demand decisions. We conclude that electricity consumption and dwellings’ and individuals’ characteristics are key determinants of the fixed charge paid by Spanish households Finally, the results point to the inefficiency of the two-part tariff as those consumers who consume more electricity pay a lower price than the others.

Which prior knowledge? Quantification of in vivo brain 13C MR spectra following 13C glucose infusion using AMARES.

The recent developments in high magnetic field 13C magnetic resonance spectroscopy with improved localization and shimming techniques have led to important gains in sensitivity and spectral resolution of 13C in vivo spectra in the rodent brain, enabling the separation of several 13C isotopomers of glutamate and glutamine. In this context, the assumptions used in spectral quantification might have a significant impact on the determination of the 13C concentrations and the related metabolic fluxes. In this study, the time domain spectral quantification algorithm AMARES (advanced method for accurate, robust and efficient spectral fitting) was applied to 13 C magnetic resonance spectroscopy spectra acquired in the rat brain at 9.4 T, following infusion of [1,6-(13)C2 ] glucose. Using both Monte Carlo simulations and in vivo data, the goal of this work was: (1) to validate the quantification of in vivo 13C isotopomers using AMARES; (2) to assess the impact of the prior knowledge on the quantification of in vivo 13C isotopomers using AMARES; (3) to compare AMARES and LCModel (linear combination of model spectra) for the quantification of in vivo 13C spectra. AMARES led to accurate and reliable 13C spectral quantification similar to those obtained using LCModel, when the frequency shifts, J-coupling constants and phase patterns of the different 13C isotopomers were included as prior knowledge in the analysis.