FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3023810035> ?p ?o ?g. }

• W3023810035 abstract "In many complex domains, the input data are often not suited for the typical vector representations used in deep learning models. For example, in knowledge representation, relational learning, and some computer vision tasks, the data are often better represented as graphs or sets. In these cases, a key challenge is to learn a representation function which is invariant to permutations of set or isomorphism of graphs. In order to handle graph isomorphism, this thesis proposes a subgraph pattern neural network with invariance to graph isomorphisms and varying local neighborhood sizes. Our key insight is to incorporate the unavoidable dependencies in the training observations of induced subgraphs into both the input features and the model architecture itself via high-order dependencies, which are still able to take node/edge labels into account and facilitate inductive reasoning. In order to learn permutation-invariant set functions, this thesis shows how the characteristics of an architecture’s computational graph impact its ability to learn in contexts with complex set dependencies, and demonstrate limitations of current methods with respect to one or more of these complexity dimensions. I also propose a new Self-Attention GRU architecture, with a computation graph that is built automatically via self-attention to minimize average interaction path lengths between set elements in the architecture’s computation graph, in order to effectively capture complex dependencies between set elements.Besides the typical set problem, a new problem of representing sets-of-sets (SoS) is proposed. In this problem, multi-level dependence and multi-level permutation invariance need to be handled jointly. To address this, I propose a hierarchical sequence attention framework (HATS) for inductive set-of-sets embeddings, and develop the stochastic optimization and inference methods required for efficient learning." @default.
• W3023810035 created "2020-05-13" @default.
• W3023810035 creator A5031033543 @default.
• W3023810035 date "2020-05-07" @default.
• W3023810035 modified "2023-09-25" @default.
• W3023810035 title "SCALABLE REPRESENTATION LEARNING WITH INVARIANCES" @default.
• W3023810035 cites W103340358 @default.
• W3023810035 cites W1568033296 @default.
• W3023810035 cites W1977970897 @default.
• W3023810035 cites W1996941442 @default.
• W3023810035 cites W2003707464 @default.
• W3023810035 cites W2008857988 @default.
• W3023810035 cites W2016753842 @default.
• W3023810035 cites W2029249040 @default.
• W3023810035 cites W2052197092 @default.
• W3023810035 cites W205829674 @default.
• W3023810035 cites W2082734581 @default.
• W3023810035 cites W2094149843 @default.
• W3023810035 cites W2101705355 @default.
• W3023810035 cites W2109480754 @default.
• W3023810035 cites W2118349699 @default.
• W3023810035 cites W2137983211 @default.
• W3023810035 cites W2142517301 @default.
• W3023810035 cites W2145544171 @default.
• W3023810035 cites W2183830099 @default.
• W3023810035 cites W2184466087 @default.
• W3023810035 cites W2184957013 @default.
• W3023810035 cites W2186482614 @default.
• W3023810035 cites W2247894828 @default.
• W3023810035 cites W2250966211 @default.
• W3023810035 cites W2401610261 @default.
• W3023810035 cites W2406128552 @default.
• W3023810035 cites W2470673105 @default.
• W3023810035 cites W2470861207 @default.
• W3023810035 cites W2515462565 @default.
• W3023810035 cites W2547418827 @default.
• W3023810035 cites W2560609797 @default.
• W3023810035 cites W2624431344 @default.
• W3023810035 cites W2785934082 @default.
• W3023810035 cites W2786776430 @default.
• W3023810035 cites W2787887656 @default.
• W3023810035 cites W2805054782 @default.
• W3023810035 cites W2805516822 @default.
• W3023810035 cites W2899038170 @default.
• W3023810035 cites W2950191616 @default.
• W3023810035 cites W2950265899 @default.
• W3023810035 cites W2952433032 @default.
• W3023810035 cites W2952689920 @default.
• W3023810035 cites W2952834907 @default.
• W3023810035 cites W2962711740 @default.
• W3023810035 cites W2962756421 @default.
• W3023810035 cites W2963084730 @default.
• W3023810035 cites W2963310665 @default.
• W3023810035 cites W2963341956 @default.
• W3023810035 cites W2963907629 @default.
• W3023810035 cites W2963982496 @default.
• W3023810035 cites W2964121744 @default.
• W3023810035 cites W2964308564 @default.
• W3023810035 cites W2970638480 @default.
• W3023810035 doi "https://doi.org/10.25394/pgs.12252347.v1" @default.
• W3023810035 hasPublicationYear "2020" @default.
• W3023810035 type Work @default.
• W3023810035 sameAs 3023810035 @default.
• W3023810035 citedByCount "0" @default.
• W3023810035 crossrefType "dissertation" @default.
• W3023810035 hasAuthorship W3023810035A5031033543 @default.
• W3023810035 hasConcept C11413529 @default.
• W3023810035 hasConcept C121332964 @default.
• W3023810035 hasConcept C131992880 @default.
• W3023810035 hasConcept C132525143 @default.
• W3023810035 hasConcept C154945302 @default.
• W3023810035 hasConcept C190470478 @default.
• W3023810035 hasConcept C203776342 @default.
• W3023810035 hasConcept C21308566 @default.
• W3023810035 hasConcept C22149727 @default.
• W3023810035 hasConcept C24890656 @default.
• W3023810035 hasConcept C33923547 @default.
• W3023810035 hasConcept C37914503 @default.
• W3023810035 hasConcept C41008148 @default.
• W3023810035 hasConcept C45374587 @default.
• W3023810035 hasConcept C48044578 @default.
• W3023810035 hasConcept C61665672 @default.
• W3023810035 hasConcept C64339825 @default.
• W3023810035 hasConcept C77088390 @default.
• W3023810035 hasConcept C80444323 @default.
• W3023810035 hasConceptScore W3023810035C11413529 @default.
• W3023810035 hasConceptScore W3023810035C121332964 @default.
• W3023810035 hasConceptScore W3023810035C131992880 @default.
• W3023810035 hasConceptScore W3023810035C132525143 @default.
• W3023810035 hasConceptScore W3023810035C154945302 @default.
• W3023810035 hasConceptScore W3023810035C190470478 @default.
• W3023810035 hasConceptScore W3023810035C203776342 @default.
• W3023810035 hasConceptScore W3023810035C21308566 @default.
• W3023810035 hasConceptScore W3023810035C22149727 @default.
• W3023810035 hasConceptScore W3023810035C24890656 @default.
• W3023810035 hasConceptScore W3023810035C33923547 @default.
• W3023810035 hasConceptScore W3023810035C37914503 @default.
• W3023810035 hasConceptScore W3023810035C41008148 @default.
• W3023810035 hasConceptScore W3023810035C45374587 @default.
• W3023810035 hasConceptScore W3023810035C48044578 @default.

• next