Thread Similarity Matrix: Visualizing Branch Divergence in GPGPU Programs

2016 ◽  
Author(s):  
Zhibin Yu ◽  
Lieven Eeckhout ◽  
Chengzhong Xu
Keyword(s):  
Similarity Matrix ◽  
Branch Divergence

scholarly journals Multi-Scale Aggregation Graph Neural Networks Based on Feature Similarity for Semi-Supervised Learning

Entropy ◽  
2021 ◽  
Vol 23 (4) ◽  
pp. 403
Author(s):  
Xun Zhang ◽  
Lanyan Yang ◽  
Bin Zhang ◽  
Ying Liu ◽  
Dong Jiang ◽  
...  
Keyword(s):  
Neural Network ◽  
Supervised Learning ◽  
Structured Data ◽  
Similarity Matrix ◽  
Scale Feature ◽  
Multi Scale ◽  
Adjacent Node ◽  
Feature Similarity ◽  
The Mean ◽  
Graph Neural Networks

The problem of extracting meaningful data through graph analysis spans a range of different fields, such as social networks, knowledge graphs, citation networks, the World Wide Web, and so on. As increasingly structured data become available, the importance of being able to effectively mine and learn from such data continues to grow. In this paper, we propose the multi-scale aggregation graph neural network based on feature similarity (MAGN), a novel graph neural network defined in the vertex domain. Our model provides a simple and general semi-supervised learning method for graph-structured data, in which only a very small part of the data is labeled as the training set. We first construct a similarity matrix by calculating the similarity of original features between all adjacent node pairs, and then generate a set of feature extractors utilizing the similarity matrix to perform multi-scale feature propagation on graphs. The output of multi-scale feature propagation is finally aggregated by using the mean-pooling operation. Our method aims to improve the model representation ability via multi-scale neighborhood aggregation based on feature similarity. Extensive experimental evaluation on various open benchmarks shows the competitive performance of our method compared to a variety of popular architectures.

Ampelographic characterization of white grapevine cul-tivars (Vitis vinifera L.) grown in Palestine

2015 ◽  
Vol 3 (1) ◽  
pp. 1-11
Author(s):  
Rezq Basheer-Salimia
Keyword(s):  
Genetic Diversity ◽  
Genetic Distance ◽  
Germplasm Conservation ◽  
Genetic Distances ◽  
Vitis Vinifera L ◽  
Similarity Matrix ◽  
Breeding Programs ◽  
Local Varieties ◽  
Synonymies And Homonymies

Abstract: In Palestine, grape culture consists of ecotypes and cultivars (also called local varieties), for which a large number of homonymous and synonymous designations exist as well as misnaming of cultivars. The present study is the first report using detailed ampelographic characterizations (39 informative traits) to assess genetic diversity and detect similarities among sixteen accessions collected from putative diverse grape genotypes In general, 30 descriptors presented highly and satisfactory divergent genotypes, whereas the remaining traits showed no or very little ampelographic variation. Based on the similarity matrix and the resulting dendrogram of these ampelographic data, distinguishable genotypes as well as some cases of synonymies and homonymies clearly exist. A synonymy case seemed to be in four genotypes including Jandali-Mfarad, Jan-dali-Mrazraz, Jandali, and Hamadani-Mattar, which indeed showed genetic distances of less than 0.5, sug-gesting their relatedness, and the possibility that they are the same genotype, but with different names. In addition, homonym cases also occur in the following pairs of “Marawi’s, Hamadani’s, and Zaini’s genotypes, in which each pair seems to be two distinctive genotypes. Finally, among the 16 examined genotypes, the Zaini-Baladi genotype tended to show the highest genetic distance values from the others and thus could be potentially incorporated into any further local or regional breeding programs as well as germplasm conservation.

Multiview Common Subspace Clustering via Coupled Low Rank Representation

2021 ◽  
Vol 12 (4) ◽  
pp. 1-25
Author(s):  
Stanley Ebhohimhen Abhadiomhen ◽  
Zhiyang Wang ◽  
Xiangjun Shen ◽  
Jianping Fan
Keyword(s):  
Common Knowledge ◽  
Subspace Clustering ◽  
Laplacian Matrix ◽  
Low Rank ◽  
Connected Components ◽  
Similarity Matrix ◽  
Benchmark Datasets ◽  
Low Rank Representation ◽  
Low Dimensional ◽  
Shared Structure

Multi-view subspace clustering (MVSC) finds a shared structure in latent low-dimensional subspaces of multi-view data to enhance clustering performance. Nonetheless, we observe that most existing MVSC methods neglect the diversity in multi-view data by considering only the common knowledge to find a shared structure either directly or by merging different similarity matrices learned for each view. In the presence of noise, this predefined shared structure becomes a biased representation of the different views. Thus, in this article, we propose a MVSC method based on coupled low-rank representation to address the above limitation. Our method first obtains a low-rank representation for each view, constrained to be a linear combination of the view-specific representation and the shared representation by simultaneously encouraging the sparsity of view-specific one. Then, it uses the k -block diagonal regularizer to learn a manifold recovery matrix for each view through respective low-rank matrices to recover more manifold structures from them. In this way, the proposed method can find an ideal similarity matrix by approximating clustering projection matrices obtained from the recovery structures. Hence, this similarity matrix denotes our clustering structure with exactly k connected components by applying a rank constraint on the similarity matrix’s relaxed Laplacian matrix to avoid spectral post-processing of the low-dimensional embedding matrix. The core of our idea is such that we introduce dynamic approximation into the low-rank representation to allow the clustering structure and the shared representation to guide each other to learn cleaner low-rank matrices that would lead to a better clustering structure. Therefore, our approach is notably different from existing methods in which the local manifold structure of data is captured in advance. Extensive experiments on six benchmark datasets show that our method outperforms 10 similar state-of-the-art compared methods in six evaluation metrics.

Computer-aided RAPD fingerprinting of accessions from the ryegrass–fescue complex

1997 ◽  
Vol 129 (3) ◽  
pp. 257-265 ◽  
Author(s):  
G. šIFFELOVÁ ◽  
M. PAVELKOVÁ ◽  
A. KLABOUCHOVÁ ◽  
I. WIESNER ◽  
V. NAšINEC
Keyword(s):  
Festuca Arundinacea ◽  
Dna Isolation ◽  
Similarity Matrix ◽  
Rapd Profile ◽  
Average Linkage ◽  
Computer Image Processing ◽  
Computer Aided ◽  
Primary Output ◽  
Taxonomic Groups ◽  
Linkage Cluster

RAPD (Randomly Amplified Polymorphic DNA) assay of 32 cultivar accessions from the ryegrass–fescue (Lolium–Festuca) complex was accomplished using ten decamer primers to assess (i) the power of RAPD technology to discriminate between individual commercial accessions and to produce cultivar fingerprinting, (ii) the degree of relatedness of accessions based on RAPD profiles in comparison with other existing classifications, and (iii) the possibility of automation of RAPD technology.The variation of the correlation coefficient r as the primary output from the automated RAPD-profile processing summarizes variability derived from DNA isolation, the RAPD reaction, and final computer-image processing of RAPD profiles. The AII (Accession Identity Interval) of r for accession Festuca arundinacea cv. Lekora was determined experimentally and the value obtained was accepted as a valid interval for all the other accessions studied. In order to evaluate the discrimination potential of all ten primers together, a pooled-similarity matrix was computed. Employing this approach, we achieved 100% discrimination between all 35 accessions when using all ten primers. A dendrogram for all 35 accessions was obtained using average linkage cluster analysis (UPGMA – Unweighted Pair Group Method with Arithmetic Means). This procedure successfully produced smaller groups of higher taxonomic homogeneity. The relationships between the Lolium–Festuca accessions were also revealed by principal coordinate analysis (PCO) based on absorbance profiles from the RAPD assay. Again, all accessions were well separated, recognising even subspecies relationships. In general, PCO analysis confirmed the inferences made from the UPGMA method.We successfully applied the computer-aided system of RAPD assay, based on an IBM PC computer, for discrimination of cultivars as well as for description of DNA-based relationships of accessions from various taxonomic groups of the Lolium–Festuca complex.

Sign in / Sign up

Export Citation Format

Share Document