1,720,956 research outputs found
Applying Natural-Language-Processing-Based Machine-Learning Techniques to our Large Scale CUDA AutoTuning Dataset
No full textAutotuning oppgaver er nesten umulige for mennesker å gjennomføre. Den abstrakte relasjonen mellom maskinvare parametere og program ytelse, gjør parameter setting uegnet for hånd. Uten autotuning, mangler programvare low-level optimaliseringer, som resulterer i mindre ytelse. Tid krevende søkemetoder går ofte hånd i hånd med autotuning. Videreføringen av maskin læring (ML) kan minske disse tidskrevende søkeprosessene.
Bruk av naturlig språk prosessering (NLP) basert ML på kildekode, for å gjennomføre autotuning oppgaver er ett voksende emne. Tidligere prosjekter har med suksess utført en rekke ulike autotuning oppgaver med flere typer kildekode språk. Mesteparten av denne kildekoden er relatert til CPU, og lite GPU kode er tilgjengelig. Med vårt LS-CAT prosjekt skapte vi ett datasett bestående av CUDA GPU kode. Denne avhandlingen implementer flere NLP-ML “pipelines” for å evaluere ML-basert «thread-coarsning» på vårt LS-CAT datasett.
Flere modell konfigurasjoner var i stand til å slå både «random choice», 0.940, og kun velge største «thread-block» størrelse (1024), 0.9437. Den beste modellen scoret 0.9483, som gir en gjennomsnittlig ytelse økning på 0.49 prosent over å velge kun den største blokken.
Avhandlingen gjorde flere oppdagelser. Implementeringen av «self-attention» mekanismer virket positivt i læringsprosessen ved å motvirke over-fitting. Valget av underliggende metodologi er viktig, og «multi-label» metoden virket best. Sammenlignet med datasettet fra tidligere forsøk, ga vårt LS-CAT datasetts høyere antall mulige «thread-coarsning» nivåer et falskt inntrykk av lavere ytelse.
Valget av «embedding” I tidligere arbeid «inst2vec» var ute av stand til å tolke rundt halvparten av «CUDA» kilde koden, som resulterte i høyt tap av data. Måter å håndtere dette og andre ideer for fremtidig arbeid er også inkludert i denne avhandlingen
Going Beyond Counting First Authors in Author Co-citation Analysis
No full textThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
No full text“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
No full textWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Applying Natural-Language-Processing-Based Machine-Learning Techniques to our Large Scale CUDA AutoTuning Dataset
No full textAutotuning oppgaver er nesten umulige for mennesker å gjennomføre. Den abstrakte relasjonen mellom maskinvare parametere og program ytelse, gjør parameter setting uegnet for hånd. Uten autotuning, mangler programvare low-level optimaliseringer, som resulterer i mindre ytelse. Tid krevende søkemetoder går ofte hånd i hånd med autotuning. Videreføringen av maskin læring (ML) kan minske disse tidskrevende søkeprosessene.
Bruk av naturlig språk prosessering (NLP) basert ML på kildekode, for å gjennomføre autotuning oppgaver er ett voksende emne. Tidligere prosjekter har med suksess utført en rekke ulike autotuning oppgaver med flere typer kildekode språk. Mesteparten av denne kildekoden er relatert til CPU, og lite GPU kode er tilgjengelig. Med vårt LS-CAT prosjekt skapte vi ett datasett bestående av CUDA GPU kode. Denne avhandlingen implementer flere NLP-ML “pipelines” for å evaluere ML-basert «thread-coarsning» på vårt LS-CAT datasett.
Flere modell konfigurasjoner var i stand til å slå både «random choice», 0.940, og kun velge største «thread-block» størrelse (1024), 0.9437. Den beste modellen scoret 0.9483, som gir en gjennomsnittlig ytelse økning på 0.49 prosent over å velge kun den største blokken.
Avhandlingen gjorde flere oppdagelser. Implementeringen av «self-attention» mekanismer virket positivt i læringsprosessen ved å motvirke over-fitting. Valget av underliggende metodologi er viktig, og «multi-label» metoden virket best. Sammenlignet med datasettet fra tidligere forsøk, ga vårt LS-CAT datasetts høyere antall mulige «thread-coarsning» nivåer et falskt inntrykk av lavere ytelse.
Valget av «embedding” I tidligere arbeid «inst2vec» var ute av stand til å tolke rundt halvparten av «CUDA» kilde koden, som resulterte i høyt tap av data. Måter å håndtere dette og andre ideer for fremtidig arbeid er også inkludert i denne avhandlingen.Autotuning tasks are almost impossible for humans to perform. The abstract relation between hardware parameters and program performance makes setting hardware parameters a far too complex task for any human. Without autotuning, software ends up missing low-level optimizations, resulting in lower performance. Traditionally time-consuming trial and error search methods have been the staple of autotuning. The emergence of machine learning (ML) could diminish these time-consuming searches.
Applying Natural language processing (NLP) based ML methods to source code as a means to perform autotuning-oriented tasks is a growing topic. The earlier projects have, with success, performed a range of different autotuning tasks using multiple source code languages. However, most of the source code data is CPU-oriented, with very little GPU code. Unsatisfied with this, our LS-CAT (Large-Scale CUDA AutoTuning) project used CUDA GPU-based kernels and generated a dataset to perform thread-coarsening. This thesis implements several custom NLP-ML pipelines to evaluate ML-based thread-coarsening using our LS-CAT dataset.
Several model configurations were able to beat both random choice, 0.9400, and the only selecting the largest thread-block (1024), 0.9437. Finally, the best model achieves a score of 0.9483, giving an average performance increase and speedup of 0.49 percent over the largest thread-block.
This project made several discoveries. The implementation of self-attention mechanisms proved beneficial in the learning process by counteracting over-fitting. The choice of underlying methodology is important, with a multi-label method outperforming the rest. Compared to the dataset from cummins, our LS-CAT dataset's higher number of thread-coarsening levels gave a false impression of lower performance.
The choice of embedding in earlier works inst2vec was unable to parse around half of the CUDA source code LLVM IR tokens, resulting in high data loss. Approaches on how to address this and other ideas for future work are also included in this thesis
Dispelling the Myths Behind First-author Citation Counts
No full textWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
koamabayili/VECTRON-author-checklist: VECTRON author checklist
No full textWe have done our best to complete the author checklist relating to the use of animals in the hut study. Note that the objective for the hut study was to evaluate the IRS treatment applications for residual efficacy against Anopheles mosquitoes, including the local An. coluzzii mosquito population. Cows were only used to attract mosquitoes into the huts and no tests were carried out directly on the cows. The author checklist is intended for use with studies where experiments are carried out on animals, which is why we have had such difficulty in completing this for the hut study, as many of the questions do not relate to how the cows were used
Author-wise bibliometric analysis based on entropy.
No full textAuthor-wise bibliometric analysis based on entropy.</p
- …
