FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3176708658> ?p ?o ?g. }

• W3176708658 abstract "Continuous deep learning architectures enable learning of flexible probabilistic models for predictive modeling as neural ordinary differential equations (ODEs), and for generative modeling as continuous normalizing flows. In this work, we design a framework to decipher the internal dynamics of these continuous depth models by pruning their network architectures. Our empirical results suggest that pruning improves generalization for neural ODEs in generative modeling. Moreover, pruning finds minimal and efficient neural ODE representations with up to 98% less parameters compared to the original network, without loss of accuracy. Finally, we show that by applying pruning we can obtain insightful information about the design of better neural ODEs.We hope our results will invigorate further research into the performance-size trade-offs of modern continuous-depth models." @default.
• W3176708658 created "2021-07-05" @default.
• W3176708658 creator A5010146439 @default.
• W3176708658 creator A5025684214 @default.
• W3176708658 creator A5064147145 @default.
• W3176708658 creator A5066830185 @default.
• W3176708658 date "2021-06-24" @default.
• W3176708658 modified "2023-09-27" @default.
• W3176708658 title "Sparse Flows: Pruning Continuous-depth Models" @default.
• W3176708658 cites W1498436455 @default.
• W3176708658 cites W1587799944 @default.
• W3176708658 cites W1821462560 @default.
• W3176708658 cites W1866230956 @default.
• W3176708658 cites W1990457366 @default.
• W3176708658 cites W2064675550 @default.
• W3176708658 cites W2113651538 @default.
• W3176708658 cites W2114766824 @default.
• W3176708658 cites W2125389748 @default.
• W3176708658 cites W2144144709 @default.
• W3176708658 cites W2409550820 @default.
• W3176708658 cites W2431962807 @default.
• W3176708658 cites W2515385951 @default.
• W3176708658 cites W2554302513 @default.
• W3176708658 cites W2619016545 @default.
• W3176708658 cites W2626325961 @default.
• W3176708658 cites W2786214536 @default.
• W3176708658 cites W2791004381 @default.
• W3176708658 cites W2912713668 @default.
• W3176708658 cites W2946795888 @default.
• W3176708658 cites W2948659923 @default.
• W3176708658 cites W2963090522 @default.
• W3176708658 cites W2963142510 @default.
• W3176708658 cites W2963287528 @default.
• W3176708658 cites W2963433148 @default.
• W3176708658 cites W2963641970 @default.
• W3176708658 cites W2963674932 @default.
• W3176708658 cites W2963755523 @default.
• W3176708658 cites W2963813662 @default.
• W3176708658 cites W2963959597 @default.
• W3176708658 cites W2964020555 @default.
• W3176708658 cites W2964233199 @default.
• W3176708658 cites W2964299589 @default.
• W3176708658 cites W2964315715 @default.
• W3176708658 cites W2964343746 @default.
• W3176708658 cites W2964833278 @default.
• W3176708658 cites W2968555978 @default.
• W3176708658 cites W2970181183 @default.
• W3176708658 cites W2970641149 @default.
• W3176708658 cites W2979652999 @default.
• W3176708658 cites W2979789219 @default.
• W3176708658 cites W2995428361 @default.
• W3176708658 cites W3005235017 @default.
• W3176708658 cites W3027429260 @default.
• W3176708658 cites W3028907532 @default.
• W3176708658 cites W3030175037 @default.
• W3176708658 cites W3034500979 @default.
• W3176708658 cites W3036415528 @default.
• W3176708658 cites W3037971676 @default.
• W3176708658 cites W3087644732 @default.
• W3176708658 cites W3090132620 @default.
• W3176708658 cites W3093985743 @default.
• W3176708658 cites W3095242670 @default.
• W3176708658 cites W3096541186 @default.
• W3176708658 cites W3100844775 @default.
• W3176708658 cites W3106103874 @default.
• W3176708658 cites W3112657725 @default.
• W3176708658 cites W3120383865 @default.
• W3176708658 cites W3132212223 @default.
• W3176708658 cites W3136258283 @default.
• W3176708658 cites W3138582970 @default.
• W3176708658 cites W3139460307 @default.
• W3176708658 cites W3172248255 @default.
• W3176708658 cites W3186412620 @default.
• W3176708658 hasPublicationYear "2021" @default.
• W3176708658 type Work @default.
• W3176708658 sameAs 3176708658 @default.
• W3176708658 citedByCount "4" @default.
• W3176708658 countsByYear W31767086582021 @default.
• W3176708658 crossrefType "posted-content" @default.
• W3176708658 hasAuthorship W3176708658A5010146439 @default.
• W3176708658 hasAuthorship W3176708658A5025684214 @default.
• W3176708658 hasAuthorship W3176708658A5064147145 @default.
• W3176708658 hasAuthorship W3176708658A5066830185 @default.
• W3176708658 hasConcept C108010975 @default.
• W3176708658 hasConcept C108583219 @default.
• W3176708658 hasConcept C11413529 @default.
• W3176708658 hasConcept C119857082 @default.
• W3176708658 hasConcept C134306372 @default.
• W3176708658 hasConcept C154945302 @default.
• W3176708658 hasConcept C167966045 @default.
• W3176708658 hasConcept C177148314 @default.
• W3176708658 hasConcept C28826006 @default.
• W3176708658 hasConcept C33923547 @default.
• W3176708658 hasConcept C34862557 @default.
• W3176708658 hasConcept C39890363 @default.
• W3176708658 hasConcept C41008148 @default.
• W3176708658 hasConcept C50644808 @default.
• W3176708658 hasConcept C51544822 @default.
• W3176708658 hasConcept C6557445 @default.
• W3176708658 hasConcept C78045399 @default.

• next