Matches in SemOpenAlex for { <https://semopenalex.org/work/W2787080188> ?p ?o ?g. }
- W2787080188 abstract "Graph learning is currently dominated by graph kernels, which, while powerful, suffer some significant limitations. Convolutional Neural Networks (CNNs) offer a very appealing alternative, but processing graphs with CNNs is not trivial. To address this challenge, many sophisticated extensions of CNNs have recently been introduced. In this paper, we reverse the problem: rather than proposing yet another graph CNN model, we introduce a novel way to represent graphs as multi-channel image-like structures that allows them to be handled by vanilla 2D CNNs. Experiments reveal that our method is more accurate than state-of-the-art graph kernels and graph CNNs on 4 out of 6 real-world datasets (with and without continuous node attributes), and close elsewhere. Our approach is also preferable to graph kernels in terms of time complexity. Code and data are publicly available." @default.
- W2787080188 created "2018-02-23" @default.
- W2787080188 creator A5011742954 @default.
- W2787080188 creator A5057695979 @default.
- W2787080188 creator A5064520344 @default.
- W2787080188 creator A5071674394 @default.
- W2787080188 date "2017-07-29" @default.
- W2787080188 modified "2023-09-27" @default.
- W2787080188 title "Graph Classification with 2D Convolutional Neural Networks" @default.
- W2787080188 cites W1522301498 @default.
- W2787080188 cites W1525595230 @default.
- W2787080188 cites W1533861849 @default.
- W2787080188 cites W1662382123 @default.
- W2787080188 cites W1816257748 @default.
- W2787080188 cites W1854214752 @default.
- W2787080188 cites W1895577753 @default.
- W2787080188 cites W1973749534 @default.
- W2787080188 cites W1999653836 @default.
- W2787080188 cites W2008857988 @default.
- W2787080188 cites W2009678996 @default.
- W2787080188 cites W2033911267 @default.
- W2787080188 cites W2095705004 @default.
- W2787080188 cites W2099438806 @default.
- W2787080188 cites W2101234009 @default.
- W2787080188 cites W2102907934 @default.
- W2787080188 cites W2108811527 @default.
- W2787080188 cites W2112796928 @default.
- W2787080188 cites W2117897510 @default.
- W2787080188 cites W2119821739 @default.
- W2787080188 cites W2124824205 @default.
- W2787080188 cites W2129905273 @default.
- W2787080188 cites W2142498761 @default.
- W2787080188 cites W2147286743 @default.
- W2787080188 cites W2158222596 @default.
- W2787080188 cites W2159156271 @default.
- W2787080188 cites W2163605009 @default.
- W2787080188 cites W2185303849 @default.
- W2787080188 cites W2242161203 @default.
- W2787080188 cites W2271840356 @default.
- W2787080188 cites W2280799770 @default.
- W2787080188 cites W229097380 @default.
- W2787080188 cites W2315403234 @default.
- W2787080188 cites W2412598452 @default.
- W2787080188 cites W2511730936 @default.
- W2787080188 cites W2519887557 @default.
- W2787080188 cites W2546945975 @default.
- W2787080188 cites W2604795503 @default.
- W2787080188 cites W2788919350 @default.
- W2787080188 cites W2950577311 @default.
- W2787080188 cites W2962756421 @default.
- W2787080188 cites W2962787457 @default.
- W2787080188 cites W2962911247 @default.
- W2787080188 cites W2964145825 @default.
- W2787080188 cites W2964321699 @default.
- W2787080188 cites W3104097132 @default.
- W2787080188 hasPublicationYear "2017" @default.
- W2787080188 type Work @default.
- W2787080188 sameAs 2787080188 @default.
- W2787080188 citedByCount "8" @default.
- W2787080188 countsByYear W27870801882017 @default.
- W2787080188 countsByYear W27870801882018 @default.
- W2787080188 countsByYear W27870801882019 @default.
- W2787080188 crossrefType "posted-content" @default.
- W2787080188 hasAuthorship W2787080188A5011742954 @default.
- W2787080188 hasAuthorship W2787080188A5057695979 @default.
- W2787080188 hasAuthorship W2787080188A5064520344 @default.
- W2787080188 hasAuthorship W2787080188A5071674394 @default.
- W2787080188 hasConcept C132525143 @default.
- W2787080188 hasConcept C153180895 @default.
- W2787080188 hasConcept C154945302 @default.
- W2787080188 hasConcept C41008148 @default.
- W2787080188 hasConcept C80444323 @default.
- W2787080188 hasConcept C81363708 @default.
- W2787080188 hasConceptScore W2787080188C132525143 @default.
- W2787080188 hasConceptScore W2787080188C153180895 @default.
- W2787080188 hasConceptScore W2787080188C154945302 @default.
- W2787080188 hasConceptScore W2787080188C41008148 @default.
- W2787080188 hasConceptScore W2787080188C80444323 @default.
- W2787080188 hasConceptScore W2787080188C81363708 @default.
- W2787080188 hasLocation W27870801881 @default.
- W2787080188 hasOpenAccess W2787080188 @default.
- W2787080188 hasPrimaryLocation W27870801881 @default.
- W2787080188 hasRelatedWork W2725194812 @default.
- W2787080188 hasRelatedWork W2770717124 @default.
- W2787080188 hasRelatedWork W2899246819 @default.
- W2787080188 hasRelatedWork W2899379687 @default.
- W2787080188 hasRelatedWork W2899886700 @default.
- W2787080188 hasRelatedWork W2935184916 @default.
- W2787080188 hasRelatedWork W2938624153 @default.
- W2787080188 hasRelatedWork W2945848398 @default.
- W2787080188 hasRelatedWork W2962965968 @default.
- W2787080188 hasRelatedWork W2963066159 @default.
- W2787080188 hasRelatedWork W2964145825 @default.
- W2787080188 hasRelatedWork W2964321699 @default.
- W2787080188 hasRelatedWork W2978348252 @default.
- W2787080188 hasRelatedWork W3015628811 @default.
- W2787080188 hasRelatedWork W3033565791 @default.
- W2787080188 hasRelatedWork W3047391881 @default.
- W2787080188 hasRelatedWork W3105136071 @default.
- W2787080188 hasRelatedWork W3118559873 @default.