FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3095883070> ?p ?o ?g. }

• W3095883070 abstract "How to obtain informative representations of molecules is a crucial prerequisite in AI-driven drug design and discovery. Recent researches abstract molecules as graphs and employ Graph Neural Networks (GNNs) for molecular representation learning. Nevertheless, two issues impede the usage of GNNs in real scenarios: (1) insufficient labeled molecules for supervised training; (2) poor generalization capability to new-synthesized molecules. To address them both, we propose a novel framework, GROVER, which stands for Graph Representation frOm self-superVised mEssage passing tRansformer. With carefully designed self-supervised tasks in node-, edge- and graph-level, GROVER can learn rich structural and semantic information of molecules from enormous unlabelled molecular data. Rather, to encode such complex information, GROVER integrates Message Passing Networks into the Transformer-style architecture to deliver a class of more expressive encoders of molecules. The flexibility of GROVER allows it to be trained efficiently on large-scale molecular dataset without requiring any supervision, thus being immunized to the two issues mentioned above. We pre-train GROVER with 100 million parameters on 10 million unlabelled molecules -- the biggest GNN and the largest training dataset in molecular representation learning. We then leverage the pre-trained GROVER for molecular property prediction followed by task-specific fine-tuning, where we observe a huge improvement (more than 6% on average) from current state-of-the-art methods on 11 challenging benchmarks. The insights we gained are that well-designed self-supervision losses and largely-expressive pre-trained models enjoy the significant potential on performance boosting." @default.
• W3095883070 created "2020-11-09" @default.
• W3095883070 creator A5005345630 @default.
• W3095883070 creator A5007120124 @default.
• W3095883070 creator A5011883201 @default.
• W3095883070 creator A5032642601 @default.
• W3095883070 creator A5045777220 @default.
• W3095883070 creator A5068865316 @default.
• W3095883070 creator A5082572337 @default.
• W3095883070 date "2020-06-18" @default.
• W3095883070 modified "2023-09-27" @default.
• W3095883070 title "Self-Supervised Graph Transformer on Large-Scale Molecular Data" @default.
• W3095883070 cites W1677182931 @default.
• W3095883070 cites W1738019091 @default.
• W3095883070 cites W1757990252 @default.
• W3095883070 cites W1988037271 @default.
• W3095883070 cites W2008505552 @default.
• W3095883070 cites W2038702914 @default.
• W3095883070 cites W2038895391 @default.
• W3095883070 cites W2060531713 @default.
• W3095883070 cites W2076498053 @default.
• W3095883070 cites W2096541451 @default.
• W3095883070 cites W2130942839 @default.
• W3095883070 cites W2145578524 @default.
• W3095883070 cites W2153693853 @default.
• W3095883070 cites W2194775991 @default.
• W3095883070 cites W2214665483 @default.
• W3095883070 cites W2279490987 @default.
• W3095883070 cites W2406943157 @default.
• W3095883070 cites W2461470610 @default.
• W3095883070 cites W2461620095 @default.
• W3095883070 cites W2473190403 @default.
• W3095883070 cites W2527189750 @default.
• W3095883070 cites W2594183968 @default.
• W3095883070 cites W2606780347 @default.
• W3095883070 cites W2619042136 @default.
• W3095883070 cites W2624431344 @default.
• W3095883070 cites W2735246657 @default.
• W3095883070 cites W2749279690 @default.
• W3095883070 cites W2766453196 @default.
• W3095883070 cites W2787998955 @default.
• W3095883070 cites W2790808809 @default.
• W3095883070 cites W2893944917 @default.
• W3095883070 cites W2895744665 @default.
• W3095883070 cites W2912078280 @default.
• W3095883070 cites W2945052571 @default.
• W3095883070 cites W2947209427 @default.
• W3095883070 cites W2962756421 @default.
• W3095883070 cites W2962767366 @default.
• W3095883070 cites W2962876364 @default.
• W3095883070 cites W2962960500 @default.
• W3095883070 cites W2963341956 @default.
• W3095883070 cites W2963403868 @default.
• W3095883070 cites W2963446712 @default.
• W3095883070 cites W2963495494 @default.
• W3095883070 cites W2963711743 @default.
• W3095883070 cites W2964015378 @default.
• W3095883070 cites W2964113829 @default.
• W3095883070 cites W2964198013 @default.
• W3095883070 cites W2965556524 @default.
• W3095883070 cites W2966357564 @default.
• W3095883070 cites W2968734407 @default.
• W3095883070 cites W2973114758 @default.
• W3095883070 cites W2991425343 @default.
• W3095883070 cites W2994710732 @default.
• W3095883070 cites W2996268457 @default.
• W3095883070 cites W2996443485 @default.
• W3095883070 cites W2999218478 @default.
• W3095883070 cites W3005552578 @default.
• W3095883070 cites W3012816161 @default.
• W3095883070 cites W3082217026 @default.
• W3095883070 cites W3104097132 @default.
• W3095883070 hasPublicationYear "2020" @default.
• W3095883070 type Work @default.
• W3095883070 sameAs 3095883070 @default.
• W3095883070 citedByCount "9" @default.
• W3095883070 countsByYear W30958830702020 @default.
• W3095883070 countsByYear W30958830702021 @default.
• W3095883070 crossrefType "posted-content" @default.
• W3095883070 hasAuthorship W3095883070A5005345630 @default.
• W3095883070 hasAuthorship W3095883070A5007120124 @default.
• W3095883070 hasAuthorship W3095883070A5011883201 @default.
• W3095883070 hasAuthorship W3095883070A5032642601 @default.
• W3095883070 hasAuthorship W3095883070A5045777220 @default.
• W3095883070 hasAuthorship W3095883070A5068865316 @default.
• W3095883070 hasAuthorship W3095883070A5082572337 @default.
• W3095883070 hasConcept C104317684 @default.
• W3095883070 hasConcept C111919701 @default.
• W3095883070 hasConcept C118505674 @default.
• W3095883070 hasConcept C119857082 @default.
• W3095883070 hasConcept C121332964 @default.
• W3095883070 hasConcept C132525143 @default.
• W3095883070 hasConcept C153083717 @default.
• W3095883070 hasConcept C154945302 @default.
• W3095883070 hasConcept C165801399 @default.
• W3095883070 hasConcept C185592680 @default.
• W3095883070 hasConcept C2780022179 @default.
• W3095883070 hasConcept C41008148 @default.
• W3095883070 hasConcept C55493867 @default.
• W3095883070 hasConcept C62520636 @default.

• next