Matches in SemOpenAlex for { <https://semopenalex.org/work/W4367693586> ?p ?o ?g. }
Showing items 1 to 49 of
49
with 100 items per page.
- W4367693586 abstract "We introduce in this paper the mechanism of graph random features (GRFs). GRFs can be used to construct unbiased randomized estimators of several important kernels defined on graphs' nodes, in particular the regularized Laplacian kernel. As regular RFs for non-graph kernels, they provide means to scale up kernel methods defined on graphs to larger networks. Importantly, they give substantial computational gains also for smaller graphs, while applied in downstream applications. Consequently, GRFs address the notoriously difficult problem of cubic (in the number of the nodes of the graph) time complexity of graph kernels algorithms. We provide a detailed theoretical analysis of GRFs and an extensive empirical evaluation: from speed tests, through Frobenius relative error analysis to kmeans graph-clustering with graph kernels. We show that the computation of GRFs admits an embarrassingly simple distributed algorithm that can be applied if the graph under consideration needs to be split across several machines. We also introduce a (still unbiased) quasi Monte Carlo variant of GRFs, q-GRFs, relying on the so-called reinforced random walks, that might be used to optimize the variance of GRFs. As a byproduct, we obtain a novel approach to solve certain classes of linear equations with positive and symmetric matrices." @default.
- W4367693586 created "2023-05-03" @default.
- W4367693586 creator A5031842812 @default.
- W4367693586 date "2023-04-28" @default.
- W4367693586 modified "2023-09-27" @default.
- W4367693586 title "Taming graph kernels with random features" @default.
- W4367693586 doi "https://doi.org/10.48550/arxiv.2305.00156" @default.
- W4367693586 hasPublicationYear "2023" @default.
- W4367693586 type Work @default.
- W4367693586 citedByCount "0" @default.
- W4367693586 crossrefType "posted-content" @default.
- W4367693586 hasAuthorship W4367693586A5031842812 @default.
- W4367693586 hasBestOaLocation W43676935861 @default.
- W4367693586 hasConcept C105795698 @default.
- W4367693586 hasConcept C11413529 @default.
- W4367693586 hasConcept C118615104 @default.
- W4367693586 hasConcept C132525143 @default.
- W4367693586 hasConcept C185429906 @default.
- W4367693586 hasConcept C33923547 @default.
- W4367693586 hasConcept C41008148 @default.
- W4367693586 hasConcept C45374587 @default.
- W4367693586 hasConcept C74193536 @default.
- W4367693586 hasConcept C80444323 @default.
- W4367693586 hasConceptScore W4367693586C105795698 @default.
- W4367693586 hasConceptScore W4367693586C11413529 @default.
- W4367693586 hasConceptScore W4367693586C118615104 @default.
- W4367693586 hasConceptScore W4367693586C132525143 @default.
- W4367693586 hasConceptScore W4367693586C185429906 @default.
- W4367693586 hasConceptScore W4367693586C33923547 @default.
- W4367693586 hasConceptScore W4367693586C41008148 @default.
- W4367693586 hasConceptScore W4367693586C45374587 @default.
- W4367693586 hasConceptScore W4367693586C74193536 @default.
- W4367693586 hasConceptScore W4367693586C80444323 @default.
- W4367693586 hasLocation W43676935861 @default.
- W4367693586 hasOpenAccess W4367693586 @default.
- W4367693586 hasPrimaryLocation W43676935861 @default.
- W4367693586 hasRelatedWork W1513831164 @default.
- W4367693586 hasRelatedWork W1568173680 @default.
- W4367693586 hasRelatedWork W1973573083 @default.
- W4367693586 hasRelatedWork W1988224349 @default.
- W4367693586 hasRelatedWork W2152704622 @default.
- W4367693586 hasRelatedWork W2166041606 @default.
- W4367693586 hasRelatedWork W2354062721 @default.
- W4367693586 hasRelatedWork W3023433514 @default.
- W4367693586 hasRelatedWork W4236163602 @default.
- W4367693586 hasRelatedWork W2944165644 @default.
- W4367693586 isParatext "false" @default.
- W4367693586 isRetracted "false" @default.
- W4367693586 workType "article" @default.