The Experts below are selected from a list of 4008 Experts worldwide ranked by ideXlab platform

Parreaux Lionel - One of the best experts on this subject based on the ideXlab platform.

  • Finally, a Polymorphic Linear Algebra Language
    2019
    Co-Authors: Shaikhha Amir, Parreaux Lionel
    Abstract:

    Many different data analytics tasks boil down to linear algebra primitives. In practice, for each different type of workload, data scientists use a particular Specialised Library. In this paper, we present Pilatus, a polymorphic iterative linear algebra language, applicable to various types of data analytics workloads. The design of this domain-specific language (DSL) is inspired by both mathematics and programming languages: its basic constructs are borrowed from abstract algebra, whereas the key technology behind its polymorphic design uses the tagless final approach (a.k.a. polymorphic embedding/object algebras). This design enables us to change the behaviour of arithmetic operations to express matrix algebra, graph algorithms, logical probabilistic programs, and differentiable programs. Crucially, the polymorphic design of Pilatus allows us to use multistage programming and rewrite-based optimisation to recover the performance of Specialised code, supporting fixed sized matrices, algebraic optimisations, and fusion

  • Finally, a Polymorphic Linear Algebra Language (Pearl)
    LIPIcs - Leibniz International Proceedings in Informatics. 33rd European Conference on Object-Oriented Programming (ECOOP 2019), 2019
    Co-Authors: Shaikhha Amir, Parreaux Lionel
    Abstract:

    Many different data analytics tasks boil down to linear algebra primitives. In practice, for each different type of workload, data scientists use a particular Specialised Library. In this paper, we present Pilatus, a polymorphic iterative linear algebra language, applicable to various types of data analytics workloads. The design of this domain-specific language (DSL) is inspired by both mathematics and programming languages: its basic constructs are borrowed from abstract algebra, whereas the key technology behind its polymorphic design uses the tagless final approach (a.k.a. polymorphic embedding/object algebras). This design enables us to change the behaviour of arithmetic operations to express matrix algebra, graph algorithms, logical probabilistic programs, and differentiable programs. Crucially, the polymorphic design of Pilatus allows us to use multi-stage programming and rewrite-based optimisation to recover the performance of Specialised code, supporting fixed sized matrices, algebraic optimisations, and fusion

Hrbek David - One of the best experts on this subject based on the ideXlab platform.

  • Users of information retrieval services in National Medical Library in Prague
    2019
    Co-Authors: Hrbek David
    Abstract:

    (in English) This bachelor thesis deals with the National Medical Library users' survey who use their search services. It shows a characteristic of the chosen users' sample and study results of their feedback. The aim of this thesis is to record interaction between National Medical Library and the users and to evaluate an asset of the search services form the Library users on the basis of collected data, respectively the fact if these users consider it to be useful for them. I have used input and output questionnaires for data collecting and the personal interviews as well. Some Specialised Library staff participated in the specific que- stions' creation. The interviews with the respondents were made by myself. A sociological profile of a common user of the search services emerged from this survey as regards to age, study or a job. Due to a specific status of the Library It could be assumed in advance that all the respondents participated in medical field. The personal interviews with the users showed to be very useful because it came out that some respondents didn't have event the basic knowledge of the literary research. This fact couldn't be seen in the written questionnaires. This bachelor thesis' topic haven't been processed too often so far. Nevertheless it was showed that this survey could..

  • Users of information retrieval services in National Medical Library in Prague
    Univerzita Karlova Filozofická fakulta, 2019
    Co-Authors: Hrbek David
    Abstract:

    (česky) Bakalářská práce se zabývá výzkumem uživatelů Národní lékařské knihovny, kteří využívají jejích rešeršních služeb. Předkládá charakteristiku vybraného vzorku uživatelů a zkoumá výsledky jejich zpětné vazby. Cílem práce je podchytit interakci mezi Národní lékařskou knihovnou a uživateli a na základě sebraných dat vyhodnotit přínos rešeršních služeb pro uživatele knihovny, respektive tu sku- tečnost, jestli uživatelé tuto službu za přínosnou považují. Ke sběru dat jsem využil vstupních a výstupních dotazníků a také osobních rozho- vorů. Na tvorbě konkrétních dotazů v dotaznících se spolupodíleli i pracovníci knihovny, zatímco osobní rozhovory jsem s respondenty vedl já sám. Z výzkumu vyplynul určitý sociologický profil běžného uživatele rešeršních slu- žeb, co se týče věku a studia či zaměstnání. Vzhledem ke specifikaci knihovny se dalo již předem předpokládat, že budou všichni respondenti zapojeni do medicín- ského či zdravotnického oboru. Velmi užitečnými se ukázaly osobní rozhovory s uživateli, neboť vyšlo najevo, že někteří z respondentů nemají ani základní po- vědomí o tom, čím rešerše je, což z dotazníků nebylo patrné. Téma této bakalářské práce není předmětem častých výzkumů a následných zpra- cování. Přesto se ukázalo, že tento výzkum by mohl být prohlouben o sběr dalších dat z jiných...(in English) This bachelor thesis deals with the National Medical Library users' survey who use their search services. It shows a characteristic of the chosen users' sample and study results of their feedback. The aim of this thesis is to record interaction between National Medical Library and the users and to evaluate an asset of the search services form the Library users on the basis of collected data, respectively the fact if these users consider it to be useful for them. I have used input and output questionnaires for data collecting and the personal interviews as well. Some Specialised Library staff participated in the specific que- stions' creation. The interviews with the respondents were made by myself. A sociological profile of a common user of the search services emerged from this survey as regards to age, study or a job. Due to a specific status of the Library It could be assumed in advance that all the respondents participated in medical field. The personal interviews with the users showed to be very useful because it came out that some respondents didn't have event the basic knowledge of the literary research. This fact couldn't be seen in the written questionnaires. This bachelor thesis' topic haven't been processed too often so far. Nevertheless it was showed that this survey could...Institute of Information Studies and LibrarianshipÚstav informačních studií a knihovnictvíFaculty of ArtsFilozofická fakult

Shaikhha Amir - One of the best experts on this subject based on the ideXlab platform.

  • Finally, a Polymorphic Linear Algebra Language
    2019
    Co-Authors: Shaikhha Amir, Parreaux Lionel
    Abstract:

    Many different data analytics tasks boil down to linear algebra primitives. In practice, for each different type of workload, data scientists use a particular Specialised Library. In this paper, we present Pilatus, a polymorphic iterative linear algebra language, applicable to various types of data analytics workloads. The design of this domain-specific language (DSL) is inspired by both mathematics and programming languages: its basic constructs are borrowed from abstract algebra, whereas the key technology behind its polymorphic design uses the tagless final approach (a.k.a. polymorphic embedding/object algebras). This design enables us to change the behaviour of arithmetic operations to express matrix algebra, graph algorithms, logical probabilistic programs, and differentiable programs. Crucially, the polymorphic design of Pilatus allows us to use multistage programming and rewrite-based optimisation to recover the performance of Specialised code, supporting fixed sized matrices, algebraic optimisations, and fusion

  • Finally, a Polymorphic Linear Algebra Language (Pearl)
    LIPIcs - Leibniz International Proceedings in Informatics. 33rd European Conference on Object-Oriented Programming (ECOOP 2019), 2019
    Co-Authors: Shaikhha Amir, Parreaux Lionel
    Abstract:

    Many different data analytics tasks boil down to linear algebra primitives. In practice, for each different type of workload, data scientists use a particular Specialised Library. In this paper, we present Pilatus, a polymorphic iterative linear algebra language, applicable to various types of data analytics workloads. The design of this domain-specific language (DSL) is inspired by both mathematics and programming languages: its basic constructs are borrowed from abstract algebra, whereas the key technology behind its polymorphic design uses the tagless final approach (a.k.a. polymorphic embedding/object algebras). This design enables us to change the behaviour of arithmetic operations to express matrix algebra, graph algorithms, logical probabilistic programs, and differentiable programs. Crucially, the polymorphic design of Pilatus allows us to use multi-stage programming and rewrite-based optimisation to recover the performance of Specialised code, supporting fixed sized matrices, algebraic optimisations, and fusion

Ross García Fernando - One of the best experts on this subject based on the ideXlab platform.

  • Retorno de la Inversión (ROI) en bibliotecas: herramientas prácticas para calcular el valor económico y social de la biblioteca
    2018
    Co-Authors: Ross García Fernando
    Abstract:

    The measurement research conducted in Spain over the last few years on the social and economic impact of libraries evidences the profitability of the services that a public, university or Specialised Library has to offer. While such research may be important, it is lacking in value unless publicised and disseminated. This is because the information about the value of libraries must be made available to society, and their social and economic profitability must be be made visible. Practical tools such as the Library services value calculator of the research done in Navarra in 2016 evinces, in a practical way, the profitability of our Library and the services we offer.Los estudios de medición del impacto económico y social de las bibliotecas realizados en España en los últimos años han demostrado la rentabilidad de los servicios que una biblioteca pública, universitaria o especializada ofrece. Si bien este tipo de estudios tienen su importantancia carecen de valor si no se publicitan y difunden, ya que es necesario hacer llegar a la sociedad la información sobre el valor de las bibliotecas y demostrar, que además de socialmente, las bibliotecas también son económicamente rentables. En este sentido las herramientas prácticas como la calculadora del valor de los servicios bibliotecarios que acompaña al estudio realizado en Navarra en 2016 permite visibilizar, de una forma práctica, la rentabilidad de nuestra biblioteca o de los servicios que ofrecemos

P. Chidambaram - One of the best experts on this subject based on the ideXlab platform.