Matches in SemOpenAlex for { <https://semopenalex.org/work/W2612865963> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W2612865963 abstract "In application domains where perceived quality is limited by human senses, where data are inherently noisy, or where models are naturally inexact, approximate computing offers an attractive tradeoff between accuracy and energy or performance. While several approximate functional units have been proposed to date, the question of how these techniques can be systematically integrated into a design flow remains open. Ideally, units like adders or multipliers could be automatically replaced with their approximate counterparts as part of the design flow. This, however, requires accurately modelling approximation errors to avoid compromising output quality. Prior proposals have either focused on describing errors per-bit or significantly limited estimation accuracy to reduce otherwise exponential storage requirements. When multiple approximate modules are chained, these limitations become critical, and propagated error estimates can be orders of magnitude off. In this paper, we propose an approach where both input distributions and approximation errors are modelled as Gaussian mixtures. This naturally represents the multiple sources of error that arise in many approximate circuits while maintaining reasonable memory requirements. Estimation accuracy is significantly better than prior art (up to 7.2× lower Hellinger distance) and errors can be accurately propagated through a cascade of approximate operations; estimates of quality metrics like MSE and MED are within a few percent of simulation-derived values." @default.
- W2612865963 created "2017-05-19" @default.
- W2612865963 creator A5058402965 @default.
- W2612865963 creator A5067207130 @default.
- W2612865963 date "2017-03-01" @default.
- W2612865963 modified "2023-09-26" @default.
- W2612865963 title "Gaussian mixture error estimation for approximate circuits" @default.
- W2612865963 cites W1965672260 @default.
- W2612865963 cites W1966502880 @default.
- W2612865963 cites W1977686543 @default.
- W2612865963 cites W1986192284 @default.
- W2612865963 cites W1996431812 @default.
- W2612865963 cites W1998824039 @default.
- W2612865963 cites W2020429547 @default.
- W2612865963 cites W2045294186 @default.
- W2612865963 cites W2045656688 @default.
- W2612865963 cites W2126044301 @default.
- W2612865963 cites W2129394555 @default.
- W2612865963 cites W2134199473 @default.
- W2612865963 cites W2147050042 @default.
- W2612865963 cites W2187230075 @default.
- W2612865963 cites W3141620748 @default.
- W2612865963 cites W3146960355 @default.
- W2612865963 cites W4231091214 @default.
- W2612865963 cites W4238042827 @default.
- W2612865963 cites W4245861455 @default.
- W2612865963 doi "https://doi.org/10.23919/date.2017.7927004" @default.
- W2612865963 hasPublicationYear "2017" @default.
- W2612865963 type Work @default.
- W2612865963 sameAs 2612865963 @default.
- W2612865963 citedByCount "3" @default.
- W2612865963 countsByYear W26128659632020 @default.
- W2612865963 countsByYear W26128659632023 @default.
- W2612865963 crossrefType "proceedings-article" @default.
- W2612865963 hasAuthorship W2612865963A5058402965 @default.
- W2612865963 hasAuthorship W2612865963A5067207130 @default.
- W2612865963 hasConcept C111472728 @default.
- W2612865963 hasConcept C11413529 @default.
- W2612865963 hasConcept C121332964 @default.
- W2612865963 hasConcept C122383733 @default.
- W2612865963 hasConcept C126255220 @default.
- W2612865963 hasConcept C134306372 @default.
- W2612865963 hasConcept C138885662 @default.
- W2612865963 hasConcept C151376022 @default.
- W2612865963 hasConcept C153024298 @default.
- W2612865963 hasConcept C163716315 @default.
- W2612865963 hasConcept C164620267 @default.
- W2612865963 hasConcept C2779530757 @default.
- W2612865963 hasConcept C28826006 @default.
- W2612865963 hasConcept C33923547 @default.
- W2612865963 hasConcept C41008148 @default.
- W2612865963 hasConcept C62520636 @default.
- W2612865963 hasConcept C76155785 @default.
- W2612865963 hasConcept C82876162 @default.
- W2612865963 hasConceptScore W2612865963C111472728 @default.
- W2612865963 hasConceptScore W2612865963C11413529 @default.
- W2612865963 hasConceptScore W2612865963C121332964 @default.
- W2612865963 hasConceptScore W2612865963C122383733 @default.
- W2612865963 hasConceptScore W2612865963C126255220 @default.
- W2612865963 hasConceptScore W2612865963C134306372 @default.
- W2612865963 hasConceptScore W2612865963C138885662 @default.
- W2612865963 hasConceptScore W2612865963C151376022 @default.
- W2612865963 hasConceptScore W2612865963C153024298 @default.
- W2612865963 hasConceptScore W2612865963C163716315 @default.
- W2612865963 hasConceptScore W2612865963C164620267 @default.
- W2612865963 hasConceptScore W2612865963C2779530757 @default.
- W2612865963 hasConceptScore W2612865963C28826006 @default.
- W2612865963 hasConceptScore W2612865963C33923547 @default.
- W2612865963 hasConceptScore W2612865963C41008148 @default.
- W2612865963 hasConceptScore W2612865963C62520636 @default.
- W2612865963 hasConceptScore W2612865963C76155785 @default.
- W2612865963 hasConceptScore W2612865963C82876162 @default.
- W2612865963 hasLocation W26128659631 @default.
- W2612865963 hasOpenAccess W2612865963 @default.
- W2612865963 hasPrimaryLocation W26128659631 @default.
- W2612865963 hasRelatedWork W2116911522 @default.
- W2612865963 hasRelatedWork W2330661264 @default.
- W2612865963 hasRelatedWork W2351491280 @default.
- W2612865963 hasRelatedWork W2371447506 @default.
- W2612865963 hasRelatedWork W2386767533 @default.
- W2612865963 hasRelatedWork W2515747357 @default.
- W2612865963 hasRelatedWork W2612865963 @default.
- W2612865963 hasRelatedWork W303980170 @default.
- W2612865963 hasRelatedWork W3141679561 @default.
- W2612865963 hasRelatedWork W4318621079 @default.
- W2612865963 isParatext "false" @default.
- W2612865963 isRetracted "false" @default.
- W2612865963 magId "2612865963" @default.
- W2612865963 workType "article" @default.