Matches in SemOpenAlex for { <https://semopenalex.org/work/W3171483544> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W3171483544 endingPage "9147" @default.
- W3171483544 startingPage "9136" @default.
- W3171483544 abstract "Latent variable models have been successfully applied in lossless compression with the bits-back coding algorithm. However, bits-back suffers from an increase in the bitrate equal to the KL divergence between the approximate posterior and the true posterior. In this paper, we show how to remove this gap asymptotically by deriving bits-back coding algorithms from tighter variational bounds. The key idea is to exploit extended space representations of Monte Carlo estimators of the marginal likelihood. Naively applied, our schemes would require more initial bits than the standard bits-back coder, but we show how to drastically reduce this additional cost with couplings in the latent space. When parallel architectures can be exploited, our coders can achieve better rates than bits-back with little additional cost. We demonstrate improved lossless compression rates in a variety of settings, especially in out-of-distribution or sequential data compression." @default.
- W3171483544 created "2021-06-22" @default.
- W3171483544 creator A5031533607 @default.
- W3171483544 creator A5046253142 @default.
- W3171483544 creator A5054711904 @default.
- W3171483544 creator A5058031547 @default.
- W3171483544 creator A5058841326 @default.
- W3171483544 creator A5074108427 @default.
- W3171483544 creator A5089038476 @default.
- W3171483544 creator A5091677854 @default.
- W3171483544 date "2021-07-18" @default.
- W3171483544 modified "2023-09-28" @default.
- W3171483544 title "Improving Lossless Compression Rates via Monte Carlo Bits-Back Coding" @default.
- W3171483544 hasPublicationYear "2021" @default.
- W3171483544 type Work @default.
- W3171483544 sameAs 3171483544 @default.
- W3171483544 citedByCount "5" @default.
- W3171483544 countsByYear W31714835442021 @default.
- W3171483544 crossrefType "proceedings-article" @default.
- W3171483544 hasAuthorship W3171483544A5031533607 @default.
- W3171483544 hasAuthorship W3171483544A5046253142 @default.
- W3171483544 hasAuthorship W3171483544A5054711904 @default.
- W3171483544 hasAuthorship W3171483544A5058031547 @default.
- W3171483544 hasAuthorship W3171483544A5058841326 @default.
- W3171483544 hasAuthorship W3171483544A5074108427 @default.
- W3171483544 hasAuthorship W3171483544A5089038476 @default.
- W3171483544 hasAuthorship W3171483544A5091677854 @default.
- W3171483544 hasConcept C105795698 @default.
- W3171483544 hasConcept C11413529 @default.
- W3171483544 hasConcept C115961682 @default.
- W3171483544 hasConcept C13481523 @default.
- W3171483544 hasConcept C153338461 @default.
- W3171483544 hasConcept C154945302 @default.
- W3171483544 hasConcept C175732694 @default.
- W3171483544 hasConcept C1769480 @default.
- W3171483544 hasConcept C19499675 @default.
- W3171483544 hasConcept C33923547 @default.
- W3171483544 hasConcept C41008148 @default.
- W3171483544 hasConcept C78548338 @default.
- W3171483544 hasConcept C80444323 @default.
- W3171483544 hasConcept C81081738 @default.
- W3171483544 hasConcept C9417928 @default.
- W3171483544 hasConcept C94835093 @default.
- W3171483544 hasConceptScore W3171483544C105795698 @default.
- W3171483544 hasConceptScore W3171483544C11413529 @default.
- W3171483544 hasConceptScore W3171483544C115961682 @default.
- W3171483544 hasConceptScore W3171483544C13481523 @default.
- W3171483544 hasConceptScore W3171483544C153338461 @default.
- W3171483544 hasConceptScore W3171483544C154945302 @default.
- W3171483544 hasConceptScore W3171483544C175732694 @default.
- W3171483544 hasConceptScore W3171483544C1769480 @default.
- W3171483544 hasConceptScore W3171483544C19499675 @default.
- W3171483544 hasConceptScore W3171483544C33923547 @default.
- W3171483544 hasConceptScore W3171483544C41008148 @default.
- W3171483544 hasConceptScore W3171483544C78548338 @default.
- W3171483544 hasConceptScore W3171483544C80444323 @default.
- W3171483544 hasConceptScore W3171483544C81081738 @default.
- W3171483544 hasConceptScore W3171483544C9417928 @default.
- W3171483544 hasConceptScore W3171483544C94835093 @default.
- W3171483544 hasOpenAccess W3171483544 @default.
- W3171483544 hasRelatedWork W140972235 @default.
- W3171483544 hasRelatedWork W1568325880 @default.
- W3171483544 hasRelatedWork W1591719784 @default.
- W3171483544 hasRelatedWork W2103588588 @default.
- W3171483544 hasRelatedWork W2343106471 @default.
- W3171483544 hasRelatedWork W275668722 @default.
- W3171483544 hasRelatedWork W2803263220 @default.
- W3171483544 hasRelatedWork W2810544457 @default.
- W3171483544 hasRelatedWork W2945490368 @default.
- W3171483544 hasRelatedWork W2946215287 @default.
- W3171483544 hasRelatedWork W2951035445 @default.
- W3171483544 hasRelatedWork W2953034051 @default.
- W3171483544 hasRelatedWork W2963948038 @default.
- W3171483544 hasRelatedWork W3019835459 @default.
- W3171483544 hasRelatedWork W3024366773 @default.
- W3171483544 hasRelatedWork W3035550066 @default.
- W3171483544 hasRelatedWork W3129867189 @default.
- W3171483544 hasRelatedWork W3161482222 @default.
- W3171483544 hasRelatedWork W3213982839 @default.
- W3171483544 hasRelatedWork W3178468572 @default.
- W3171483544 isParatext "false" @default.
- W3171483544 isRetracted "false" @default.
- W3171483544 magId "3171483544" @default.
- W3171483544 workType "article" @default.