112 research outputs found
Reducing Communication in AMG for Reservoir Simulation: Aggressive Coarsening and Non-Galerkin Coarse-Grid Operators
No full textAlgebraic Multigrid (AMG) is an efficient multigrid method for solving large problems, using only the information provided by the underlying matrices. Unfortunately, on a parallel machine the performance of AMG can be negatively affected by dense communication patterns. The exchange of extensive data sets is generally caused by a high number of non-zero entries contained by the coarse grid operators. We discuss two solution strategies, that can be applied in order to improve the sparsity patterns of these operators: aggressive coarsening and the non-Galerkin method. Aggressive coarsening reduces the number of coarse grid variables by modifying the basic concepts of the standard coarsening scheme. The non-Galerkin approach is entirely different. While preserving the row sums, the non-Galerkin method removes less important entries of the coarse level operators, generated by AMG. In fact, it can be seen as an extension of the standard multigrid algorithm. We explore both techniques in the frame of reservoir simulation. We demonstrate that aggressive coarsening and non-Galerkin algorithm significantly reduce the total number of non-zero entries in the AMG hierarchy. Although aggressive coarsening shows high performance, in terms of execution time, on one processor core, it is less effective for parallel simulations. The non-Galerkin has not been tested in parallel, but its serial results are very promissing.Applied MathematicsDelft Institute for Applied MathematicsElectrical Engineering, Mathematics and Computer Scienc
Tuning the Coarse Space Construction in a Spectral AMG Solver1
No full textAbstractIn this paper, we discuss strategies for computing subsets of eigenvectors of matrices corresponding to subdomains of finite element meshes achieving compromise between two contradicting goals. The subset of eigenvectors is required in the construction of coarse spaces used in algebraic multigrid methods (AMG) as well as in certain domain decomposition (DD) methods. The quality of the coarse spaces depends on the number of eigenvectors, which improves the approximation properties of the coarse space and impacts the overall performance and convergence of the associated AMG or DD algorithms. However, a large number of eigenvectors affects negatively the sparsity of the corresponding coarse matrices, which can become fairly dense. The sparsity of the coarse matrices can be controlled to a certain extent by the size of the subdomains (union of finite elements) referred to as agglomerates. If the size of the agglomerates is too large, then the cost of the eigensolvers increases and eventually can become unacceptable for the purpose of constructing the AMG or DD solvers. This paper investigates strategies to optimize the solution of the partial eigenproblems of interest. In particular, we examine direct and iterative eigensolvers for computing those subsets. Our experiments with a well-known model of an oil-reservoir simulation benchmark indicate that iterative eigensolvers can lead to significant improvements in the overall performance of an AMG solver that exploits such spectral construction of coarse spaces
Numerical Methods for Large Thermo-Mechanical Systems
No full textNumerical methods are investigated for solving large-scale sparse linear systems of equations, that can be applied to thermo-mechanical models and wafer-slip models. This thesis examines efficient numerical methods, in terms of memory, number of iterations required for convergence, and computation time. To be more specific, algebraic multigrid (AMG) methods and deflation methods are considered as preconditioners for the conjugate gradient method. We investigate if smoothed aggregation AMG or adaptive smoothing and prolongation based AMG improve upon the classical Ruge-Stüben AMG. It is shown that Ruge-Stüben AMG needs fewer iterations for the test problems. However, smoothed aggregation AMG has a smaller data-size, which is of interest for situations with limited memory or large systems of equations. Moreover, the mechanical problems considered have a coefficient matrix with a block structure, which can be exploited by preconditioners like block Jacobi or the incomplete block Cholesky decomposition; but also the smoothed aggregation AMG can take the block structure into account when creating coarser grids. Further, we examine if the results of the conjugate gradient method can be improved by adding a deflation preconditioner based on the proper orthogonal decomposition or rigid body modes. They are combined with a direct or stationary iterative preconditioner, resulting in two-level preconditioned conjugate gradient methods. The various implementations of such methods are discussed, and the deflation preconditioner is shown to generally reduce the number of iterations compared to the single preconditioner.Applied Mathematic
Applying multi-label techniques in emotion identification of short texts
No full textSentiment Analysis is an emerging research field traditionally applied to classify opinions, sentiments and emotions towards polarity and subjectivity expressed in text. An important characteristic to automatic emotion analysis is the standpoint, in which we can look at an opinion from two perspectives, the opinion holder (author) who express an opinion, and the reader who reads and perceives the opinion. From the reader's standpoint, the interpretations of the text can be multiple and depend on the personal background. The multiple standpoints cognition, in which readers can look at the same sentence, is an interesting scenario to use the multi-label classification paradigm in the Sentiment Analysis domain. This methodology is able to handle different target sentiments simultaneously in the same text, by also taking advantage of the relations between them. We applied different approaches such as algorithm adaptation, problem transformation and ensemble methods in order to explore the wide range of multi-label solutions. The experiments were conducted on 10,080 news sentences from two different real datasets. Experimental results showed that the Ensemble Classifier Chain overcame the other algorithms, average F-measure of 64.89% using emotion strength features, when considering six emotions and neutral sentiment. (C) 2018 Elsevier B.V. All rights reserved
Further on a contribution to establish the understanding of §§ 40, 41 AMG
No full text0\. Titelseite, Danksagung, Zusammenfassung
Abstract, Gliederung, Literaturverzeichnis
1\. Einleitung und Darstellung des Problems 1
2\. Mögliche Rechtsgrundlagen für die Placebokontrolle im AMG 5
3\. Vereinbarkeit von § 41 II, III AMG mit Art. 1 I, 2 II, 3 I und 6 I
Grundgesetz sowie Art. 7 S. 2 des Internationalen Paktes für bürgerliche und
politische Rechte 153
4\. Vorschläge für eine teilweise Neuformulierung des AMG 198
5\. Zusammenfassung in Thesenform 204Die Möglichkeit von Placebokontrollen bei der klinischen Prüfung am Menschen
gemäß Arzneimittelgesetz (AMG) hat auch nach dessen im Juli 2004 ergangener
12. Gesetzesnovelle keinerlei explizite Regelung erfahren. Im Rahmen der
vorliegenden Arbeit wird der Umfang der rechtlichen Zulässigkeit
placebokontrollierter Forschung anhand der neuen gesetzlichen Vorschriften
erstmals detailliert untersucht. Im Zentrum der Untersuchung steht zunächst
die Frage, inwieweit sich die Vorschriften des Arzneimittelgesetzes über
klinische Prüfungen überhaupt als placebo-relevant darstellen. Die für die
Placebokontrolle als Rechtsgrundlagen in Betracht kommenden Vorschriften des
Arzneimittelgesetzes über klinische Prüfungen am Menschen werden dabei unter
Berücksichtigung der durch die 12. Gesetzesnovelle umgesetzten europäischen
Richtlinien, insbesondere 2001/83/EG und 2001/20/EG, der Biomedizin-Konvention
des Europarates von 1997 und der Deklaration von Helsinki des Weltärztebundes
sowie der Verordnung über die Anwendung der Guten Klinischen Praxis GCP-VO
interpretiert. Es wird herausgearbeitet, dass die Zuführung eines
Placebopräparates grundsätzlich dem Regelungsbereich des Arzneimittelgesetzes
unterfällt was mangels ausdrücklicher Erwähnung indessen bezweifelt werden
könnte , nicht hingegen allgemeine Regeln für die Beurteilung der rechtlichen
Zulässigkeit heranzuziehen sind. Dabei stehen sowohl die Bestimmung des
Arzneimittelbegriffs als auch die Definition der klinischen Prüfung im Zentrum
der Betrachtungen. Nach Beantwortung dieser Vorfrage werden die für eine
Placebogabe in der Kontrollgruppe in Betracht kommenden Rechtsgrundlagen
erörtert. Diese sind innerhalb der Vorschriften der §§ 40, 41 AMG zu suchen,
welche die allgemeinen und besonderen Voraussetzungen der klinischen Prüfung
am Menschen detailliert regeln. Im Rahmen der §§ 40, 41 AMG können insgesamt
fünf verschiedene Arten von Teilnehmern an klinischen Prüfungen unterschieden
werden, für die jeweils besondere Voraussetzungen gelten und dementsprechend
auch für die Placebo-Gabe gelten könnten: gesunde einwilligungsfähige
Volljährige, kranke einwilligungsfähige Volljährige, kranke einwilligungs-
unfähige Volljährige, gesunde Minderjährige und kranke Minderjährige. Die
unterschiedlichen Anforderungen an die Zulässigkeit der Placeboverabreichung
an die Prüfungsteilnehmer werden im Hinblick auf jede Teilnehmergruppe
ausführlich dargestellt und erörtert. Nachdem das Gemeinschaftsrecht und auch
das Recht des Europarates im Rahmen der Gesetzes-auslegung schon mehrfach zur
Verdeutlichung des nationalrechtlichen Befundes herangezogen werden, erfolgt
im Anschluss an die Gesetzesauslegung eine Überprüfung speziell des § 41 Abs.
2 und 3 AMG auf verfassungsrechtliche und völkerrechtliche Zulässigkeit hin.
Beim Verfassungs-recht geht es um Art. 1 Abs. 1, 2 Abs. 2, 3 Abs. 1 und 6 Abs.
2 des Grundgesetzes, beim Völker-recht um Art. 7 insbesondere S. 2 des
Internationalen Paktes über bürgerliche und politische Rechte. Abschließend
werden auf der Grundlage der Arbeitsergebnisse Neuformulierungen des
Arzneimittelgesetzes vorgeschlagen, auf deren Basis die bisherige Rechtslage
klarer und sicherer gestaltet werden könnte.The possibility to control for placebo when undertaking clinical examinations
of human beings has not been explicitly regulated within the AMG (German
pharmaceutical law). Even after the 12th amendment from July 2004, no explicit
regulation has taken place. This thesis intends to analyse for the first time
in detail the legitimacy of clinical examinations controlled for placebo and
its coverage under the new law. Initially, the analysis is focused on the
question whether or not the regulations of the pharmaceutical law on clinical
examinations are relevant for placebo testing. This is done by an
interpretation of relevant regulations within the pharmaceutical law on
clinical examinations of human beings. For the interpretation, the author
considers in particular European directives (especially 2001/83/EG &
2001/20/EG), the Biomedical Convention of the European Council from 1997 and
the declaration of Helsinki by the World Medical Association (WMA) as well as
the Regulation on Good Clinical Practice GCP-VO which have been implemented
into German law by the 12th Amendment. It is elaborated, that the usage of
placebo supplements in general is covered by the jurisdiction of
pharmaceutical law. This result could have been doubted since placebo
supplements are not explicitly mentioned. However, general rules for the
legitimacy assessment are not to be used. The analysis is centred around the
concept of pharmaceuticals as well as the definition of clinical examination.
After answering the initial question, the legal foundations for placebo
testing within the control group are examined. These can be found among §§ 40,
41 AMG, which regulate in detail the general and the specific preconditions
for clinical examinations of human beings. Within §§ 40, 41 AMG, five
different types of participants for clinical examinations can be
distinguished. For each type, different preconditions apply and thus could be
applied for placebo medication, too. The five types are: healthy adults of
full age able to give consent; sick adults of full age able to give consent,
sick adults of full age unable to give consent, healthy minors and sick
minors. The different preconditions for legitimacy of placebo medication with
reference to the different participant groups are analysed and examined in
detail. Having used Community law and the law of the European Council to
clarify the interpretation of the national law, a further examination is
afterwards undertaken. This examination intends to assess in particular § 41
Abs. 2 und 3 AMG relating to its constitutional legitimacy and its legitimacy
under international law. In constitutional law, articles1 para.1, 2 para.2, 3
para.1 and 6 para.2 of the Grundgesetz are concerned, in international law,
article 7 (in particular p.2) of the international pact on civil and political
rights is concerned. Conclusively, new formulations of the pharmaceutical law
are suggested on the basis of the undertaken results. These could help to
reform the current legal foundation and to make it more secure
The Effects of Open Data Communities on Open Data Benefits and Barriers
No full textGovernmental organisations worldwide are working on their open data strategies and are openly sharing their data with the public. Recently also open data communities have been set up by several national governments. The effects of setting up these data communities on open data benefits and barriers and whether an effect exists at all is not yet clear, because there has been little scientific research into data communities. This thesis therefore focuses on the effects of setting up open data communities. A single-unit case study was conducted to gather qualitive information of the effects of data communities. The case study focussed on the Dutch open data community and consisted of two parts, a document analysis and in-depth interviews. The results of the analysis show how the Dutch community contributes to enhancing open data benefits, including (indirectly) creating more informed citizens, increasing the access to capacity and resources outside of the data publishing organisation and a higher problem-solving capacity. Furthermore, the interview participants agree to a large extent that communities contribute to intragovernmental collaboration and the use of collective intelligence to solve public problems. According to the interviewees, the community also (potentially) mitigates open data barriers such as the lack of interest in using open data (by governmental organisations). The interviewees stated that the community managers made sure every question that was posted got a sufficient answer within a reasonable amount of time and therefore the barrier stating that the data provider ignores requests and suggestions of data users could also be mitigated, as well as difficulties in the interaction with the data provider. Both researchers and most community users and managers argued that the community could also contribute to mitigating low engagement of public managers with open data and increasing the knowledge and skills of employees to use the open data. Lastly, according to the interview participants, the community can decrease difficulty in discovering/locating data and not being able to combine and connect datasets. The interviewees were also questioned about how institutional instruments could increase the value of the open data community. Although the participants concluded that the contribution of formal instruments (such as rules) is limited, they indicated informal rules (such as norms) and enforcing instruments (such as rewards) can contribute to the value that is created by an open data community. The community-specific challenges identified in the analysis can contribute to the process of designing an open data community, because policy makers can compare scope and design choices with the empirical experiences. This thesis also provides an exploratory scientific contribution: this research provides insight into how open data communities can contribute to enhancing open data benefits and the mitigation of open data barriers. Although the results are promising, certain limitations are applicable to the research. The thesis only studied one open data community and the full list of open data benefits and barriers was reduced based upon an assessment of the author of the thesis. Last of all, the interviews focused on qualitative data. Combining the results of this thesis with other case studies can improve generalisability. Complex Systems Engineering and Management (CoSEM
Impact van C-ITS use cases: Verslag challenge, Den Haag, 11 juni 2019
No full textCoöperatieve intelligente transportsystemen (C-ITS) zijn volop in ontwikkeling. Dit kan een grote impact hebben op het Nederlandse wegennet. Rijkswaterstaat (RWS) werkt binnen het programma C-ITS Next aan opschaling en uitrol van kansrijke C-ITS diensten. Hierbij onderzoekt Rijkswater-staat welke mogelijkheden er zijn om toepassingen binnen een periode van twee jaar te realiseren en welke effecten dit kan hebben. In het kader hiervan is een workshop gehouden waarbij experts van Rijkswaterstaat samen met een aantal experts van TNO op het gebied van C-ITS samengeko-men zijn om vijf diensten te bespreken en een eerste inschatting van de effecten te maken, als opzet voor verder onderzoek. Het ging daarbij om een inschatting van de effecten op doorstro-ming, veiligheid en duurzaamheid en om een inventarisatie van de randvoorwaarden en nevenef-fecten.Transport and Plannin
An Introduction to Algebraic Multigrid An Introduction to Algebraic Multigrid
No full textAbstract Algebraic multigrid (AMG) solves linear systems based on multigrid principles, but in a way that only depends on the coefficients in the underlying matrix. The author begins with a basic introduction to AMG methods, and then describes some more recent advances and theoretical developments
Recommended from our members
An Introduction to Algebraic Multigrid
No full textAlgebraic multigrid (AMG) solves linear systems based on multigrid principles, but in a way that only depends on the coefficients in the underlying matrix. The author begins with a basic introduction to AMG methods, and then describes some more recent advances and theoretical development
Enhancing Business Data Sharing in the Supply Chain Domain: A Framework of Infrastructural and Institutional Instruments
No full textEffective data sharing plays a pivotal role in optimizing supply chain management and driving the operational excellence of businesses. However, certain barriers and challenges exist among the supply chain partners in communication and data exchange. This thesis investigates these barriers and proposes potential solutions, referred to as instruments, for data sharing in supply chains. The objective is to develop and validate a framework that addresses feasible barriers in high-tech supply chain organizations. Single case study interviews in a high-tech supply chain firm assess the framework's applicability, followed by a validation phase to test its generalizability. Infrastructural instruments include ICT, Blockchain, AI, ML, data standardization, and data security measures. Institutional instruments encompass cultural factors, trust-building techniques, contractual agreements, education and training programs, leadership practices, and ethical data sharing. The research contribution extends to a strategic deployment plan for these instruments, offering valuable insights to supply chain professionals. Additionally, the research emphasizes the significance of viewing risk as a comprehensive concept in relation to trust, technology, privacy, and governance-related barriers. Management of Technology (MoT
- …
