FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4306392318> ?p ?o ?g. }

• W4306392318 abstract "Graph Convolutional Neural Network (GCNN) is a popular class of deep learning (DL) models in material science to predict material properties from the graph representation of molecular structures. Training an accurate and comprehensive GCNN surrogate for molecular design requires large-scale graph datasets and is usually a time-consuming process. Recent advances in GPUs and distributed computing open a path to reduce the computational cost for GCNN training effectively. However, efficient utilization of high performance computing (HPC) resources for training requires simultaneously optimizing large-scale data management and scalable stochastic batched optimization techniques. In this work, we focus on building GCNN models on HPC systems to predict material properties of millions of molecules. We use HydraGNN, our in-house library for large-scale GCNN training, leveraging distributed data parallelism in PyTorch. We use ADIOS, a high-performance data management framework for efficient storage and reading of large molecular graph data. We perform parallel training on two open-source large-scale graph datasets to build a GCNN predictor for an important quantum property known as the HOMO-LUMO gap. We measure the scalability, accuracy, and convergence of our approach on two DOE supercomputers: the Summit supercomputer at the Oak Ridge Leadership Computing Facility (OLCF) and the Perlmutter system at the National Energy Research Scientific Computing Center (NERSC). We present our experimental results with HydraGNN showing (i) reduction of data loading time up to 4.2 times compared with a conventional method and (ii) linear scaling performance for training up to 1024 GPUs on both Summit and Perlmutter." @default.
• W4306392318 created "2022-10-17" @default.
• W4306392318 creator A5019108731 @default.
• W4306392318 creator A5019868556 @default.
• W4306392318 creator A5025779441 @default.
• W4306392318 creator A5054451484 @default.
• W4306392318 creator A5060842906 @default.
• W4306392318 date "2022-10-17" @default.
• W4306392318 modified "2023-09-26" @default.
• W4306392318 title "Scalable training of graph convolutional neural networks for fast and accurate predictions of HOMO-LUMO gap in molecules" @default.
• W4306392318 cites W1563453094 @default.
• W4306392318 cites W1975147762 @default.
• W4306392318 cites W1992156271 @default.
• W4306392318 cites W2032756564 @default.
• W4306392318 cites W2046001409 @default.
• W4306392318 cites W2080635178 @default.
• W4306392318 cites W2116341502 @default.
• W4306392318 cites W2337110853 @default.
• W4306392318 cites W2478294658 @default.
• W4306392318 cites W2613900957 @default.
• W4306392318 cites W2766856748 @default.
• W4306392318 cites W2779187024 @default.
• W4306392318 cites W2888395196 @default.
• W4306392318 cites W2949095042 @default.
• W4306392318 cites W2988070836 @default.
• W4306392318 cites W3007517495 @default.
• W4306392318 cites W3041193038 @default.
• W4306392318 cites W3094681328 @default.
• W4306392318 cites W3117551848 @default.
• W4306392318 cites W3119512242 @default.
• W4306392318 cites W3126702256 @default.
• W4306392318 cites W3135446204 @default.
• W4306392318 cites W3179469670 @default.
• W4306392318 cites W3180364792 @default.
• W4306392318 cites W3201073812 @default.
• W4306392318 cites W3212512279 @default.
• W4306392318 cites W4200578858 @default.
• W4306392318 cites W4206938161 @default.
• W4306392318 cites W4225312033 @default.
• W4306392318 cites W4225842713 @default.
• W4306392318 cites W4281249911 @default.
• W4306392318 cites W4281925462 @default.
• W4306392318 cites W4286899761 @default.
• W4306392318 doi "https://doi.org/10.1186/s13321-022-00652-1" @default.
• W4306392318 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/36253845" @default.
• W4306392318 hasPublicationYear "2022" @default.
• W4306392318 type Work @default.
• W4306392318 citedByCount "2" @default.
• W4306392318 countsByYear W43063923182023 @default.
• W4306392318 crossrefType "journal-article" @default.
• W4306392318 hasAuthorship W4306392318A5019108731 @default.
• W4306392318 hasAuthorship W4306392318A5019868556 @default.
• W4306392318 hasAuthorship W4306392318A5025779441 @default.
• W4306392318 hasAuthorship W4306392318A5054451484 @default.
• W4306392318 hasAuthorship W4306392318A5060842906 @default.
• W4306392318 hasBestOaLocation W43063923181 @default.
• W4306392318 hasConcept C108583219 @default.
• W4306392318 hasConcept C124101348 @default.
• W4306392318 hasConcept C130318100 @default.
• W4306392318 hasConcept C132525143 @default.
• W4306392318 hasConcept C13280743 @default.
• W4306392318 hasConcept C154945302 @default.
• W4306392318 hasConcept C173608175 @default.
• W4306392318 hasConcept C185798385 @default.
• W4306392318 hasConcept C205649164 @default.
• W4306392318 hasConcept C41008148 @default.
• W4306392318 hasConcept C48044578 @default.
• W4306392318 hasConcept C75684735 @default.
• W4306392318 hasConcept C77088390 @default.
• W4306392318 hasConcept C80444323 @default.
• W4306392318 hasConcept C81363708 @default.
• W4306392318 hasConcept C83283714 @default.
• W4306392318 hasConceptScore W4306392318C108583219 @default.
• W4306392318 hasConceptScore W4306392318C124101348 @default.
• W4306392318 hasConceptScore W4306392318C130318100 @default.
• W4306392318 hasConceptScore W4306392318C132525143 @default.
• W4306392318 hasConceptScore W4306392318C13280743 @default.
• W4306392318 hasConceptScore W4306392318C154945302 @default.
• W4306392318 hasConceptScore W4306392318C173608175 @default.
• W4306392318 hasConceptScore W4306392318C185798385 @default.
• W4306392318 hasConceptScore W4306392318C205649164 @default.
• W4306392318 hasConceptScore W4306392318C41008148 @default.
• W4306392318 hasConceptScore W4306392318C48044578 @default.
• W4306392318 hasConceptScore W4306392318C75684735 @default.
• W4306392318 hasConceptScore W4306392318C77088390 @default.
• W4306392318 hasConceptScore W4306392318C80444323 @default.
• W4306392318 hasConceptScore W4306392318C81363708 @default.
• W4306392318 hasConceptScore W4306392318C83283714 @default.
• W4306392318 hasFunder F4320306084 @default.
• W4306392318 hasIssue "1" @default.
• W4306392318 hasLocation W43063923181 @default.
• W4306392318 hasLocation W43063923182 @default.
• W4306392318 hasLocation W43063923183 @default.
• W4306392318 hasLocation W43063923184 @default.
• W4306392318 hasLocation W43063923185 @default.
• W4306392318 hasLocation W43063923186 @default.
• W4306392318 hasLocation W43063923187 @default.
• W4306392318 hasOpenAccess W4306392318 @default.
• W4306392318 hasPrimaryLocation W43063923181 @default.
• W4306392318 hasRelatedWork W112744582 @default.

• next