Matches in SemOpenAlex for { <https://semopenalex.org/work/W4282004119> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W4282004119 abstract "Rationale is defined as a subset of input features that best explains or supports the prediction by machine learning models. Rationale identification has improved the generalizability and interpretability of neural networks on vision and language data. In graph applications such as molecule and polymer property prediction, identifying representative subgraph structures named as graph rationales plays an essential role in the performance of graph neural networks. Existing graph pooling and/or distribution intervention methods suffer from the lack of examples to learn to identify optimal graph rationales. In this work, we introduce a new augmentation operation called environment replacement that automatically creates virtual data examples to improve rationale identification. We propose an efficient framework that performs rationale-environment separation and representation learning on the real and augmented examples in latent spaces to avoid the high complexity of explicit graph decoding and encoding. Comparing against recent techniques, experiments on seven molecular and four polymer datasets demonstrate the effectiveness and efficiency of the proposed augmentation-based graph rationalization framework. Data and the implementation of the proposed framework are publicly available https://github.com/liugangcode/GREA." @default.
- W4282004119 created "2022-06-13" @default.
- W4282004119 creator A5035766567 @default.
- W4282004119 creator A5065387614 @default.
- W4282004119 creator A5068746442 @default.
- W4282004119 creator A5074821819 @default.
- W4282004119 creator A5087343858 @default.
- W4282004119 date "2022-08-14" @default.
- W4282004119 modified "2023-10-02" @default.
- W4282004119 title "Graph Rationalization with Environment-based Augmentations" @default.
- W4282004119 cites W2076902316 @default.
- W4282004119 cites W2092368765 @default.
- W4282004119 cites W2883528235 @default.
- W4282004119 cites W2954839056 @default.
- W4282004119 cites W2998496395 @default.
- W4282004119 cites W3035739162 @default.
- W4282004119 cites W3080371803 @default.
- W4282004119 cites W3080644389 @default.
- W4282004119 cites W3090892334 @default.
- W4282004119 cites W3100721696 @default.
- W4282004119 cites W3114168740 @default.
- W4282004119 cites W3159527717 @default.
- W4282004119 cites W3197057534 @default.
- W4282004119 cites W3198839083 @default.
- W4282004119 cites W3208180646 @default.
- W4282004119 cites W4200158407 @default.
- W4282004119 cites W4214578516 @default.
- W4282004119 doi "https://doi.org/10.1145/3534678.3539347" @default.
- W4282004119 hasPublicationYear "2022" @default.
- W4282004119 type Work @default.
- W4282004119 citedByCount "3" @default.
- W4282004119 countsByYear W42820041192023 @default.
- W4282004119 crossrefType "proceedings-article" @default.
- W4282004119 hasAuthorship W4282004119A5035766567 @default.
- W4282004119 hasAuthorship W4282004119A5065387614 @default.
- W4282004119 hasAuthorship W4282004119A5068746442 @default.
- W4282004119 hasAuthorship W4282004119A5074821819 @default.
- W4282004119 hasAuthorship W4282004119A5087343858 @default.
- W4282004119 hasBestOaLocation W42820041192 @default.
- W4282004119 hasConcept C105795698 @default.
- W4282004119 hasConcept C119857082 @default.
- W4282004119 hasConcept C124101348 @default.
- W4282004119 hasConcept C132525143 @default.
- W4282004119 hasConcept C154945302 @default.
- W4282004119 hasConcept C27158222 @default.
- W4282004119 hasConcept C2781067378 @default.
- W4282004119 hasConcept C33923547 @default.
- W4282004119 hasConcept C41008148 @default.
- W4282004119 hasConcept C70437156 @default.
- W4282004119 hasConcept C80444323 @default.
- W4282004119 hasConceptScore W4282004119C105795698 @default.
- W4282004119 hasConceptScore W4282004119C119857082 @default.
- W4282004119 hasConceptScore W4282004119C124101348 @default.
- W4282004119 hasConceptScore W4282004119C132525143 @default.
- W4282004119 hasConceptScore W4282004119C154945302 @default.
- W4282004119 hasConceptScore W4282004119C27158222 @default.
- W4282004119 hasConceptScore W4282004119C2781067378 @default.
- W4282004119 hasConceptScore W4282004119C33923547 @default.
- W4282004119 hasConceptScore W4282004119C41008148 @default.
- W4282004119 hasConceptScore W4282004119C70437156 @default.
- W4282004119 hasConceptScore W4282004119C80444323 @default.
- W4282004119 hasFunder F4320306076 @default.
- W4282004119 hasLocation W42820041191 @default.
- W4282004119 hasLocation W42820041192 @default.
- W4282004119 hasLocation W42820041193 @default.
- W4282004119 hasOpenAccess W4282004119 @default.
- W4282004119 hasPrimaryLocation W42820041191 @default.
- W4282004119 hasRelatedWork W3006943036 @default.
- W4282004119 hasRelatedWork W4200511449 @default.
- W4282004119 hasRelatedWork W4206534706 @default.
- W4282004119 hasRelatedWork W4229079080 @default.
- W4282004119 hasRelatedWork W4285141508 @default.
- W4282004119 hasRelatedWork W4312515694 @default.
- W4282004119 hasRelatedWork W4366769859 @default.
- W4282004119 hasRelatedWork W4385957992 @default.
- W4282004119 hasRelatedWork W4385965371 @default.
- W4282004119 hasRelatedWork W4386025632 @default.
- W4282004119 isParatext "false" @default.
- W4282004119 isRetracted "false" @default.
- W4282004119 workType "article" @default.