FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2950352605> ?p ?o ?g. }

• W2950352605 abstract "Knowledge Graph (KG) embedding is a fundamental problem in data mining research with many real-world applications. It aims to encode the entities and relations in the graph into low dimensional vector space, which can be used for subsequent algorithms. Negative sampling, which samples negative triplets from non-observed ones in the training data, is an important step in KG embedding. Recently, generative adversarial network (GAN), has been introduced in negative sampling. By sampling negative triplets with large scores, these methods avoid the problem of vanishing gradient and thus obtain better performance. However, using GAN makes the original model more complex and hard to train, where reinforcement learning must be used. In this paper, motivated by the observation that negative triplets with large scores are important but rare, we propose to directly keep track of them with the cache. However, how to sample from and update the cache are two important questions. We carefully design the solutions, which are not only efficient but also achieve a good balance between exploration and exploitation. In this way, our method acts as a distilled version of previous GA-based methods, which does not waste training time on additional parameters to fit the full distribution of negative triplets. The extensive experiments show that our method can gain significant improvement in various KG embedding models, and outperform the state-of-the-art negative sampling methods based on GAN." @default.
• W2950352605 created "2019-06-27" @default.
• W2950352605 creator A5010738468 @default.
• W2950352605 creator A5014615052 @default.
• W2950352605 creator A5045112676 @default.
• W2950352605 creator A5072484211 @default.
• W2950352605 date "2018-12-16" @default.
• W2950352605 modified "2023-09-26" @default.
• W2950352605 title "NSCaching: Simple and Efficient Negative Sampling for Knowledge Graph Embedding" @default.
• W2950352605 cites W102708294 @default.
• W2950352605 cites W1522301498 @default.
• W2950352605 cites W1529533208 @default.
• W2950352605 cites W1533230146 @default.
• W2950352605 cites W1533861849 @default.
• W2950352605 cites W1585529040 @default.
• W2950352605 cites W1756422141 @default.
• W2950352605 cites W1801721664 @default.
• W2950352605 cites W1894439495 @default.
• W2950352605 cites W2022166150 @default.
• W2950352605 cites W2055317074 @default.
• W2950352605 cites W205829674 @default.
• W2950352605 cites W2081580037 @default.
• W2950352605 cites W2094728533 @default.
• W2950352605 cites W2099471712 @default.
• W2950352605 cites W2119717200 @default.
• W2950352605 cites W2125027602 @default.
• W2950352605 cites W2125891623 @default.
• W2950352605 cites W2127795553 @default.
• W2950352605 cites W2131571251 @default.
• W2950352605 cites W2132984949 @default.
• W2950352605 cites W2141599568 @default.
• W2950352605 cites W2145544171 @default.
• W2950352605 cites W2152790380 @default.
• W2950352605 cites W2172684358 @default.
• W2950352605 cites W2184957013 @default.
• W2950352605 cites W2250184916 @default.
• W2950352605 cites W2250342289 @default.
• W2950352605 cites W2283196293 @default.
• W2950352605 cites W2296073425 @default.
• W2950352605 cites W2432356473 @default.
• W2950352605 cites W2433281745 @default.
• W2950352605 cites W2487770199 @default.
• W2950352605 cites W2759136286 @default.
• W2950352605 cites W2788170399 @default.
• W2950352605 cites W2804078698 @default.
• W2950352605 cites W2809001617 @default.
• W2950352605 cites W2896255488 @default.
• W2950352605 cites W2899771611 @default.
• W2950352605 cites W2962756421 @default.
• W2950352605 cites W2962793481 @default.
• W2950352605 cites W2962879692 @default.
• W2950352605 cites W2963485453 @default.
• W2950352605 cites W2963534707 @default.
• W2950352605 cites W2964140943 @default.
• W2950352605 cites W68132019 @default.
• W2950352605 doi "https://doi.org/10.48550/arxiv.1812.06410" @default.
• W2950352605 hasPublicationYear "2018" @default.
• W2950352605 type Work @default.
• W2950352605 sameAs 2950352605 @default.
• W2950352605 citedByCount "2" @default.
• W2950352605 countsByYear W29503526052019 @default.
• W2950352605 countsByYear W29503526052020 @default.
• W2950352605 crossrefType "posted-content" @default.
• W2950352605 hasAuthorship W2950352605A5010738468 @default.
• W2950352605 hasAuthorship W2950352605A5014615052 @default.
• W2950352605 hasAuthorship W2950352605A5045112676 @default.
• W2950352605 hasAuthorship W2950352605A5072484211 @default.
• W2950352605 hasBestOaLocation W29503526051 @default.
• W2950352605 hasConcept C104317684 @default.
• W2950352605 hasConcept C106131492 @default.
• W2950352605 hasConcept C111472728 @default.
• W2950352605 hasConcept C11413529 @default.
• W2950352605 hasConcept C115537543 @default.
• W2950352605 hasConcept C119857082 @default.
• W2950352605 hasConcept C132525143 @default.
• W2950352605 hasConcept C138885662 @default.
• W2950352605 hasConcept C140779682 @default.
• W2950352605 hasConcept C154945302 @default.
• W2950352605 hasConcept C173608175 @default.
• W2950352605 hasConcept C185592680 @default.
• W2950352605 hasConcept C2780586882 @default.
• W2950352605 hasConcept C31972630 @default.
• W2950352605 hasConcept C41008148 @default.
• W2950352605 hasConcept C41608201 @default.
• W2950352605 hasConcept C55493867 @default.
• W2950352605 hasConcept C66746571 @default.
• W2950352605 hasConcept C80444323 @default.
• W2950352605 hasConcept C97541855 @default.
• W2950352605 hasConceptScore W2950352605C104317684 @default.
• W2950352605 hasConceptScore W2950352605C106131492 @default.
• W2950352605 hasConceptScore W2950352605C111472728 @default.
• W2950352605 hasConceptScore W2950352605C11413529 @default.
• W2950352605 hasConceptScore W2950352605C115537543 @default.
• W2950352605 hasConceptScore W2950352605C119857082 @default.
• W2950352605 hasConceptScore W2950352605C132525143 @default.
• W2950352605 hasConceptScore W2950352605C138885662 @default.
• W2950352605 hasConceptScore W2950352605C140779682 @default.
• W2950352605 hasConceptScore W2950352605C154945302 @default.
• W2950352605 hasConceptScore W2950352605C173608175 @default.
• W2950352605 hasConceptScore W2950352605C185592680 @default.

• next