Matches in SemOpenAlex for { <https://semopenalex.org/work/W3087685042> ?p ?o ?g. }
- W3087685042 abstract "Recently program learning techniques have been proposed to process source code based on syntactical structures (e.g., Abstract Syntax Trees) and/or semantic information (e.g., Dependency Graphs). Although graphs may be better at capturing various viewpoints of code semantics than trees, constructing graph inputs from code needs static code semantic analysis that may not be accurate and introduces noise during learning. Although syntax trees are precisely defined according to the language grammar and easier to construct and process than graphs, previous tree-based learning techniques have not been able to learn semantic information from trees to achieve better accuracy than graph-based techniques. We propose a new learning technique, named TreeCaps, by fusing together capsule networks with tree-based convolutional neural networks, to achieve learning accuracy higher than existing graph-based techniques while it is based only on trees. TreeCaps introduces novel variable-to-static routing algorithms into the capsule networks to compensate for the loss of previous routing algorithms. Aside from accuracy, we also find that TreeCaps is the most robust to withstand those semantic-preserving program transformations that change code syntax without modifying the semantics. Evaluated on a large number of Java and C/C++ programs, TreeCaps models outperform prior deep learning models of program source code, in terms of both accuracy and robustness for program comprehension tasks such as code functionality classification and function name prediction" @default.
- W3087685042 created "2020-09-25" @default.
- W3087685042 creator A5002289161 @default.
- W3087685042 creator A5033649567 @default.
- W3087685042 creator A5083048049 @default.
- W3087685042 date "2020-09-05" @default.
- W3087685042 modified "2023-09-25" @default.
- W3087685042 title "TreeCaps: Tree-Based Capsule Networks for Source Code Processing" @default.
- W3087685042 cites W1423339008 @default.
- W3087685042 cites W1545528966 @default.
- W3087685042 cites W1553894716 @default.
- W3087685042 cites W1655956671 @default.
- W3087685042 cites W1784932861 @default.
- W3087685042 cites W1966948031 @default.
- W3087685042 cites W2079887492 @default.
- W3087685042 cites W2089735638 @default.
- W3087685042 cites W2735043505 @default.
- W3087685042 cites W2741561716 @default.
- W3087685042 cites W2743316948 @default.
- W3087685042 cites W2785994986 @default.
- W3087685042 cites W2806718802 @default.
- W3087685042 cites W2809662498 @default.
- W3087685042 cites W2884276923 @default.
- W3087685042 cites W2887364112 @default.
- W3087685042 cites W2890194927 @default.
- W3087685042 cites W2909882940 @default.
- W3087685042 cites W2921355642 @default.
- W3087685042 cites W2950898568 @default.
- W3087685042 cites W2954380329 @default.
- W3087685042 cites W2955426500 @default.
- W3087685042 cites W2961882312 @default.
- W3087685042 cites W2962894772 @default.
- W3087685042 cites W2962961569 @default.
- W3087685042 cites W2962986764 @default.
- W3087685042 cites W2963015915 @default.
- W3087685042 cites W2963355447 @default.
- W3087685042 cites W2963371736 @default.
- W3087685042 cites W2963393993 @default.
- W3087685042 cites W2963499994 @default.
- W3087685042 cites W2963703618 @default.
- W3087685042 cites W2964150020 @default.
- W3087685042 cites W2965481041 @default.
- W3087685042 cites W2968117147 @default.
- W3087685042 cites W2970186524 @default.
- W3087685042 cites W2970862273 @default.
- W3087685042 cites W2994689640 @default.
- W3087685042 cites W2996086147 @default.
- W3087685042 cites W2997451752 @default.
- W3087685042 cites W3004658838 @default.
- W3087685042 cites W3006165421 @default.
- W3087685042 cites W3016881068 @default.
- W3087685042 cites W3047099251 @default.
- W3087685042 cites W3101228802 @default.
- W3087685042 cites W3106437163 @default.
- W3087685042 doi "https://doi.org/10.48550/arxiv.2009.09777" @default.
- W3087685042 hasPublicationYear "2020" @default.
- W3087685042 type Work @default.
- W3087685042 sameAs 3087685042 @default.
- W3087685042 citedByCount "0" @default.
- W3087685042 crossrefType "posted-content" @default.
- W3087685042 hasAuthorship W3087685042A5002289161 @default.
- W3087685042 hasAuthorship W3087685042A5033649567 @default.
- W3087685042 hasAuthorship W3087685042A5083048049 @default.
- W3087685042 hasBestOaLocation W30876850421 @default.
- W3087685042 hasConcept C119857082 @default.
- W3087685042 hasConcept C149091818 @default.
- W3087685042 hasConcept C154945302 @default.
- W3087685042 hasConcept C199360897 @default.
- W3087685042 hasConcept C204321447 @default.
- W3087685042 hasConcept C2777561058 @default.
- W3087685042 hasConcept C2777904410 @default.
- W3087685042 hasConcept C41008148 @default.
- W3087685042 hasConcept C43126263 @default.
- W3087685042 hasConcept C58646249 @default.
- W3087685042 hasConcept C60048249 @default.
- W3087685042 hasConcept C80444323 @default.
- W3087685042 hasConceptScore W3087685042C119857082 @default.
- W3087685042 hasConceptScore W3087685042C149091818 @default.
- W3087685042 hasConceptScore W3087685042C154945302 @default.
- W3087685042 hasConceptScore W3087685042C199360897 @default.
- W3087685042 hasConceptScore W3087685042C204321447 @default.
- W3087685042 hasConceptScore W3087685042C2777561058 @default.
- W3087685042 hasConceptScore W3087685042C2777904410 @default.
- W3087685042 hasConceptScore W3087685042C41008148 @default.
- W3087685042 hasConceptScore W3087685042C43126263 @default.
- W3087685042 hasConceptScore W3087685042C58646249 @default.
- W3087685042 hasConceptScore W3087685042C60048249 @default.
- W3087685042 hasConceptScore W3087685042C80444323 @default.
- W3087685042 hasLocation W30876850421 @default.
- W3087685042 hasLocation W30876850422 @default.
- W3087685042 hasLocation W30876850423 @default.
- W3087685042 hasOpenAccess W3087685042 @default.
- W3087685042 hasPrimaryLocation W30876850421 @default.
- W3087685042 hasRelatedWork W1507167368 @default.
- W3087685042 hasRelatedWork W2013531410 @default.
- W3087685042 hasRelatedWork W2017163896 @default.
- W3087685042 hasRelatedWork W2253889033 @default.
- W3087685042 hasRelatedWork W2359482398 @default.
- W3087685042 hasRelatedWork W270927586 @default.
- W3087685042 hasRelatedWork W2808670711 @default.