FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3216158204> ?p ?o ?g. }

• W3216158204 abstract "Most set prediction models in deep learning use set-equivariant operations, but they actually operate on multisets. We show that set-equivariant functions cannot represent certain functions on multisets, so we introduce the more appropriate notion of multiset-equivariance. We identify that the existing Deep Set Prediction Network (DSPN) can be multiset-equivariant without being hindered by set-equivariance and improve it with approximate implicit differentiation, allowing for better optimization while being faster and saving memory. In a range of toy experiments, we show that the perspective of multiset-equivariance is beneficial and that our changes to DSPN achieve better results in most cases. On CLEVR object property prediction, we substantially improve over the state-of-the-art Slot Attention from 8% to 77% in one of the strictest evaluation metrics because of the benefits made possible by implicit differentiation." @default.
• W3216158204 created "2021-12-06" @default.
• W3216158204 creator A5024508073 @default.
• W3216158204 creator A5037060625 @default.
• W3216158204 creator A5051076050 @default.
• W3216158204 creator A5060327141 @default.
• W3216158204 creator A5086664284 @default.
• W3216158204 date "2021-11-23" @default.
• W3216158204 modified "2023-09-24" @default.
• W3216158204 title "Multiset-Equivariant Set Prediction with Approximate Implicit Differentiation" @default.
• W3216158204 cites W104184427 @default.
• W3216158204 cites W1542343170 @default.
• W3216158204 cites W196761320 @default.
• W3216158204 cites W1978642336 @default.
• W3216158204 cites W2031342017 @default.
• W3216158204 cites W2046033161 @default.
• W3216158204 cites W2194775991 @default.
• W3216158204 cites W2222512263 @default.
• W3216158204 cites W2277195237 @default.
• W3216158204 cites W2505728881 @default.
• W3216158204 cites W2561715562 @default.
• W3216158204 cites W2579549467 @default.
• W3216158204 cites W2784996692 @default.
• W3216158204 cites W2786103815 @default.
• W3216158204 cites W2913668534 @default.
• W3216158204 cites W2952433032 @default.
• W3216158204 cites W2953273646 @default.
• W3216158204 cites W2962711740 @default.
• W3216158204 cites W2963403868 @default.
• W3216158204 cites W2963716836 @default.
• W3216158204 cites W2963717490 @default.
• W3216158204 cites W2963907629 @default.
• W3216158204 cites W2963970238 @default.
• W3216158204 cites W2964051675 @default.
• W3216158204 cites W2964076774 @default.
• W3216158204 cites W2964108670 @default.
• W3216158204 cites W2964121744 @default.
• W3216158204 cites W2969489725 @default.
• W3216158204 cites W2970638480 @default.
• W3216158204 cites W2970697704 @default.
• W3216158204 cites W2970791107 @default.
• W3216158204 cites W2970900903 @default.
• W3216158204 cites W2970971581 @default.
• W3216158204 cites W2972949456 @default.
• W3216158204 cites W2994687091 @default.
• W3216158204 cites W3035462522 @default.
• W3216158204 cites W3035752085 @default.
• W3216158204 cites W3040538742 @default.
• W3216158204 cites W3046819526 @default.
• W3216158204 cites W3096609285 @default.
• W3216158204 cites W3098967720 @default.
• W3216158204 cites W3099723694 @default.
• W3216158204 cites W3102506330 @default.
• W3216158204 cites W3102696055 @default.
• W3216158204 cites W3104801438 @default.
• W3216158204 cites W3129524796 @default.
• W3216158204 cites W3167630251 @default.
• W3216158204 cites W3169203126 @default.
• W3216158204 cites W3180196270 @default.
• W3216158204 cites W3181414820 @default.
• W3216158204 cites W3201167453 @default.
• W3216158204 cites W3204483762 @default.
• W3216158204 cites W3208667955 @default.
• W3216158204 cites W607748843 @default.
• W3216158204 doi "https://doi.org/10.48550/arxiv.2111.12193" @default.
• W3216158204 hasPublicationYear "2021" @default.
• W3216158204 type Work @default.
• W3216158204 sameAs 3216158204 @default.
• W3216158204 citedByCount "0" @default.
• W3216158204 crossrefType "posted-content" @default.
• W3216158204 hasAuthorship W3216158204A5024508073 @default.
• W3216158204 hasAuthorship W3216158204A5037060625 @default.
• W3216158204 hasAuthorship W3216158204A5051076050 @default.
• W3216158204 hasAuthorship W3216158204A5060327141 @default.
• W3216158204 hasAuthorship W3216158204A5086664284 @default.
• W3216158204 hasBestOaLocation W32161582041 @default.
• W3216158204 hasConcept C111472728 @default.
• W3216158204 hasConcept C11413529 @default.
• W3216158204 hasConcept C118615104 @default.
• W3216158204 hasConcept C138885662 @default.
• W3216158204 hasConcept C154945302 @default.
• W3216158204 hasConcept C159985019 @default.
• W3216158204 hasConcept C171036898 @default.
• W3216158204 hasConcept C177264268 @default.
• W3216158204 hasConcept C189950617 @default.
• W3216158204 hasConcept C192562407 @default.
• W3216158204 hasConcept C199360897 @default.
• W3216158204 hasConcept C202444582 @default.
• W3216158204 hasConcept C204323151 @default.
• W3216158204 hasConcept C2779623528 @default.
• W3216158204 hasConcept C2781238097 @default.
• W3216158204 hasConcept C33923547 @default.
• W3216158204 hasConcept C41008148 @default.
• W3216158204 hasConcept C48103436 @default.
• W3216158204 hasConcept C80444323 @default.
• W3216158204 hasConceptScore W3216158204C111472728 @default.
• W3216158204 hasConceptScore W3216158204C11413529 @default.
• W3216158204 hasConceptScore W3216158204C118615104 @default.
• W3216158204 hasConceptScore W3216158204C138885662 @default.
• W3216158204 hasConceptScore W3216158204C154945302 @default.

• next