Results 301 to 310 of about 2,949,927 (358)

FPGA Acceleration of GCN in Light of the Symmetry of Graph Adjacency Matrix

Design, Automation and Test in Europe, 2023
Graph Convolutional Neural Networks (GCNs) are widely used to process large-scale graph data. Different from deep neural networks (DNNs), GCNs are sparse, irregular, and unstructured, posing unique challenges to hardware acceleration with regular ...
Gopikrishnan Raveendran Nair, Han-Sok Suh, M. Halappanavar, Frank Liu, J.-s. Seo, Yu Cao   +5 more
semanticscholar   +1 more source

Keyword Extraction Using Unsupervised Learning on the Document’s Adjacency Matrix

Proceedings of the Fifteenth Workshop on Graph-Based Methods for Natural Language Processing (TextGraphs-15), 2021
This work revisits the information given by the graph-of-words and its typical utilization through graph-based ranking approaches in the context of keyword extraction.
Eirini Papagiannopoulou, Grigorios Tsoumakas, A. Papadopoulos   +2 more
semanticscholar   +1 more source

Pairwise Adjacency Matrix on Spatial Temporal Graph Convolution Network for Skeleton-Based Two-Person Interaction Recognition

International Conference on Information Photonics, 2020
Spatial-temporal graph convolutional networks (ST-GCN) have achieved outstanding performances on human action recognition, however, it might be less superior on a two-person interaction recognition (TPIR) task due to the relationship of each skeleton is ...
Chao-Lung Yang, Aji Setyoko, Hendrik Tampubolon, K. Hua   +3 more
semanticscholar   +1 more source

Optical Adjacency Matrix Associative Processor

SPIE Proceedings, 1990
A new bidirectional optical associative processor is described for searching a hierarchical database that is stored as an adjacency matrix. The paper discusses how the processor can answer relatively complex queries on a knowledge base when the queries are formulated as combinations of set closures, unions, intersections, and complementations.
David Casasent, Brian Telfer
openaire   +1 more source

Adjacency matrix and its square

Graphs and Geometry, 2019
openaire   +2 more sources

Bounds for the spectral radius and energy of extended adjacency matrix of graphs

Linear and multilinear algebra, 2019
An extend adjacency matrix of a graph ( ) was introduced decades ago as a precursor for developing a few quite useful molecular topological descriptors. The spectral radius ( ) of the extended adjacency matrix and the extended energy of a graph ( ) have ...
Zhao Wang, Y. Mao, Boris Furtula, Xu Wang   +3 more
semanticscholar   +1 more source

An FPGA Implementation of GCN with Sparse Adjacency Matrix

International Conference on ASIC, 2019
Deep learning has revolutionized many machine learning tasks in recent years, ranging from image classification and video processing to speech recognition and natural language understanding.
Luchang Ding, Zhize Huang, Gengsheng Chen   +2 more
semanticscholar   +1 more source

Two new topological indices based on graph adjacency matrix eigenvalues and eigenvectors

Journal of Mathematical Chemistry, 2019
The Estrada topological index EE, based on the eigenvalues of the adjacency matrix, is degenerate for cospectral graphs. By additionally considering the eigenvectors, two new topological indices are devised, which have reduced degeneracy for alkanes or ...
J. A. Rodríguez-Velázquez, A. Balaban
semanticscholar   +1 more source

Bounds for eigenvalues of the adjacency matrix of a graph

Journal of Interdisciplinary Mathematics, 2019
We obtain bounds for the largest and least eigenvalues of the adjacency matrix of a simple undirected graph. We find upper bound for the second largest eigenvalue of the adjacency matrix.
Pintu Bhunia, Santanu Bag, K. Paul
semanticscholar   +1 more source

A Compact Form of the Adjacency Matrix

Journal of Chemical Information and Computer Sciences, 2000
It has been shown that the adjacency matrix can be transformed into a row vector and then into a single number. This number can again be decoded to recover the row vector, and this in turn can be decoded to restore the original adjacency matrix. A special, rather efficient coding scheme was devised for acyclic structures.
openaire   +2 more sources