Matches in SemOpenAlex for { <https://semopenalex.org/work/W4382129596> ?p ?o ?g. }
- W4382129596 endingPage "952" @default.
- W4382129596 startingPage "938" @default.
- W4382129596 abstract "We explore different strategies to integrate prior domain knowledge into the design of graph neural networks (GNN). Our study is supported by a use-case of estimating the potential energy of chemical systems (molecules and crystals) represented as graphs. We integrate two elements of domain knowledge into the design of the GNN to constrain and regularise its learning, towards higher accuracy and generalisation. First, knowledge on the existence of different types of relations/graph edges (e.g. chemical bonds in our case study) between nodes of the graph is used to modulate their interactions. We formulate and compare two strategies, namely specialised message production and specialised update of internal states. Second, knowledge of the relevance of some physical quantities is used to constrain the learnt features towards a higher physical relevance using a simple multi-task learning (MTL) paradigm. We explore the potential of MTL to better capture the underlying mechanisms behind the studied phenomenon. We demonstrate the general applicability of our two knowledge integrations by applying them to three architectures that rely on different mechanisms to propagate information between nodes and to update node states. Our implementations are made publicly available. To support these experiments, we release three new datasets of out-of-equilibrium molecules and crystals of various complexities." @default.
- W4382129596 created "2023-06-27" @default.
- W4382129596 creator A5039077302 @default.
- W4382129596 creator A5048472742 @default.
- W4382129596 creator A5048551572 @default.
- W4382129596 date "2023-08-01" @default.
- W4382129596 modified "2023-10-07" @default.
- W4382129596 title "Domain-informed graph neural networks: A quantum chemistry case study" @default.
- W4382129596 cites W1980170509 @default.
- W4382129596 cites W2080635178 @default.
- W4382129596 cites W2116341502 @default.
- W4382129596 cites W2153693853 @default.
- W4382129596 cites W2526511911 @default.
- W4382129596 cites W2527189750 @default.
- W4382129596 cites W2585152223 @default.
- W4382129596 cites W2604314403 @default.
- W4382129596 cites W2752435843 @default.
- W4382129596 cites W2753709519 @default.
- W4382129596 cites W2902474924 @default.
- W4382129596 cites W2923693308 @default.
- W4382129596 cites W2963677766 @default.
- W4382129596 cites W2965857891 @default.
- W4382129596 cites W2968734407 @default.
- W4382129596 cites W2971323980 @default.
- W4382129596 cites W2983997668 @default.
- W4382129596 cites W3016218187 @default.
- W4382129596 cites W3102449990 @default.
- W4382129596 cites W3106310231 @default.
- W4382129596 cites W3129238601 @default.
- W4382129596 cites W3152893301 @default.
- W4382129596 cites W3212512279 @default.
- W4382129596 cites W4210257598 @default.
- W4382129596 cites W4225405705 @default.
- W4382129596 doi "https://doi.org/10.1016/j.neunet.2023.06.030" @default.
- W4382129596 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37453397" @default.
- W4382129596 hasPublicationYear "2023" @default.
- W4382129596 type Work @default.
- W4382129596 citedByCount "0" @default.
- W4382129596 crossrefType "journal-article" @default.
- W4382129596 hasAuthorship W4382129596A5039077302 @default.
- W4382129596 hasAuthorship W4382129596A5048472742 @default.
- W4382129596 hasAuthorship W4382129596A5048551572 @default.
- W4382129596 hasBestOaLocation W43821295962 @default.
- W4382129596 hasConcept C119857082 @default.
- W4382129596 hasConcept C132525143 @default.
- W4382129596 hasConcept C134306372 @default.
- W4382129596 hasConcept C154945302 @default.
- W4382129596 hasConcept C158154518 @default.
- W4382129596 hasConcept C17744445 @default.
- W4382129596 hasConcept C178790620 @default.
- W4382129596 hasConcept C185592680 @default.
- W4382129596 hasConcept C199360897 @default.
- W4382129596 hasConcept C199539241 @default.
- W4382129596 hasConcept C207685749 @default.
- W4382129596 hasConcept C26713055 @default.
- W4382129596 hasConcept C2991951333 @default.
- W4382129596 hasConcept C32909587 @default.
- W4382129596 hasConcept C33923547 @default.
- W4382129596 hasConcept C36503486 @default.
- W4382129596 hasConcept C41008148 @default.
- W4382129596 hasConcept C50644808 @default.
- W4382129596 hasConcept C80444323 @default.
- W4382129596 hasConceptScore W4382129596C119857082 @default.
- W4382129596 hasConceptScore W4382129596C132525143 @default.
- W4382129596 hasConceptScore W4382129596C134306372 @default.
- W4382129596 hasConceptScore W4382129596C154945302 @default.
- W4382129596 hasConceptScore W4382129596C158154518 @default.
- W4382129596 hasConceptScore W4382129596C17744445 @default.
- W4382129596 hasConceptScore W4382129596C178790620 @default.
- W4382129596 hasConceptScore W4382129596C185592680 @default.
- W4382129596 hasConceptScore W4382129596C199360897 @default.
- W4382129596 hasConceptScore W4382129596C199539241 @default.
- W4382129596 hasConceptScore W4382129596C207685749 @default.
- W4382129596 hasConceptScore W4382129596C26713055 @default.
- W4382129596 hasConceptScore W4382129596C2991951333 @default.
- W4382129596 hasConceptScore W4382129596C32909587 @default.
- W4382129596 hasConceptScore W4382129596C33923547 @default.
- W4382129596 hasConceptScore W4382129596C36503486 @default.
- W4382129596 hasConceptScore W4382129596C41008148 @default.
- W4382129596 hasConceptScore W4382129596C50644808 @default.
- W4382129596 hasConceptScore W4382129596C80444323 @default.
- W4382129596 hasLocation W43821295961 @default.
- W4382129596 hasLocation W43821295962 @default.
- W4382129596 hasLocation W43821295963 @default.
- W4382129596 hasLocation W43821295964 @default.
- W4382129596 hasLocation W43821295965 @default.
- W4382129596 hasOpenAccess W4382129596 @default.
- W4382129596 hasPrimaryLocation W43821295961 @default.
- W4382129596 hasRelatedWork W2002978035 @default.
- W4382129596 hasRelatedWork W2120447654 @default.
- W4382129596 hasRelatedWork W2128223750 @default.
- W4382129596 hasRelatedWork W2144453115 @default.
- W4382129596 hasRelatedWork W2188872161 @default.
- W4382129596 hasRelatedWork W2209382646 @default.
- W4382129596 hasRelatedWork W2586133538 @default.
- W4382129596 hasRelatedWork W4238532390 @default.
- W4382129596 hasRelatedWork W797688974 @default.
- W4382129596 hasRelatedWork W2977179488 @default.