Matches in SemOpenAlex for { <https://semopenalex.org/work/W3138678423> ?p ?o ?g. }
Showing items 1 to 98 of
98
with 100 items per page.
- W3138678423 abstract "Classification performance based on ImageNet is the de-facto standard metric for CNN development. In this work we challenge the notion that CNN architecture design solely based on ImageNet leads to generally effective convolutional neural network (CNN) architectures that perform well on a diverse set of datasets and application domains. To this end, we investigate and ultimately improve ImageNet as a basis for deriving such architectures. We conduct an extensive empirical study for which we train $500$ CNN architectures, sampled from the broad AnyNetX design space, on ImageNet as well as $8$ additional well known image classification benchmark datasets from a diverse array of application domains. We observe that the performances of the architectures are highly dataset dependent. Some datasets even exhibit a negative error correlation with ImageNet across all architectures. We show how to significantly increase these correlations by utilizing ImageNet subsets restricted to fewer classes. These contributions can have a profound impact on the way we design future CNN architectures and help alleviate the tilt we see currently in our community with respect to over-reliance on one dataset." @default.
- W3138678423 created "2021-03-29" @default.
- W3138678423 creator A5021570324 @default.
- W3138678423 creator A5029576638 @default.
- W3138678423 creator A5044629673 @default.
- W3138678423 date "2022-11-15" @default.
- W3138678423 modified "2023-10-14" @default.
- W3138678423 title "Is it enough to optimize CNN architectures on ImageNet?" @default.
- W3138678423 cites W1689711448 @default.
- W3138678423 cites W1903029394 @default.
- W3138678423 cites W1992396639 @default.
- W3138678423 cites W2064675550 @default.
- W3138678423 cites W2097117768 @default.
- W3138678423 cites W2141125852 @default.
- W3138678423 cites W2145607950 @default.
- W3138678423 cites W2165803222 @default.
- W3138678423 cites W2752782242 @default.
- W3138678423 cites W2884916014 @default.
- W3138678423 cites W2911779594 @default.
- W3138678423 cites W2952020226 @default.
- W3138678423 cites W2963716183 @default.
- W3138678423 cites W2964081807 @default.
- W3138678423 cites W2997849610 @default.
- W3138678423 cites W3095166428 @default.
- W3138678423 cites W3102785203 @default.
- W3138678423 cites W4297775537 @default.
- W3138678423 doi "https://doi.org/10.3389/fcomp.2022.1041703" @default.
- W3138678423 hasPublicationYear "2022" @default.
- W3138678423 type Work @default.
- W3138678423 sameAs 3138678423 @default.
- W3138678423 citedByCount "3" @default.
- W3138678423 countsByYear W31386784232021 @default.
- W3138678423 countsByYear W31386784232022 @default.
- W3138678423 countsByYear W31386784232023 @default.
- W3138678423 crossrefType "journal-article" @default.
- W3138678423 hasAuthorship W3138678423A5021570324 @default.
- W3138678423 hasAuthorship W3138678423A5029576638 @default.
- W3138678423 hasAuthorship W3138678423A5044629673 @default.
- W3138678423 hasBestOaLocation W31386784231 @default.
- W3138678423 hasConcept C115961682 @default.
- W3138678423 hasConcept C119857082 @default.
- W3138678423 hasConcept C123657996 @default.
- W3138678423 hasConcept C13280743 @default.
- W3138678423 hasConcept C142362112 @default.
- W3138678423 hasConcept C153349607 @default.
- W3138678423 hasConcept C154945302 @default.
- W3138678423 hasConcept C162324750 @default.
- W3138678423 hasConcept C176217482 @default.
- W3138678423 hasConcept C177264268 @default.
- W3138678423 hasConcept C185798385 @default.
- W3138678423 hasConcept C199360897 @default.
- W3138678423 hasConcept C205649164 @default.
- W3138678423 hasConcept C21547014 @default.
- W3138678423 hasConcept C41008148 @default.
- W3138678423 hasConcept C75294576 @default.
- W3138678423 hasConcept C81363708 @default.
- W3138678423 hasConceptScore W3138678423C115961682 @default.
- W3138678423 hasConceptScore W3138678423C119857082 @default.
- W3138678423 hasConceptScore W3138678423C123657996 @default.
- W3138678423 hasConceptScore W3138678423C13280743 @default.
- W3138678423 hasConceptScore W3138678423C142362112 @default.
- W3138678423 hasConceptScore W3138678423C153349607 @default.
- W3138678423 hasConceptScore W3138678423C154945302 @default.
- W3138678423 hasConceptScore W3138678423C162324750 @default.
- W3138678423 hasConceptScore W3138678423C176217482 @default.
- W3138678423 hasConceptScore W3138678423C177264268 @default.
- W3138678423 hasConceptScore W3138678423C185798385 @default.
- W3138678423 hasConceptScore W3138678423C199360897 @default.
- W3138678423 hasConceptScore W3138678423C205649164 @default.
- W3138678423 hasConceptScore W3138678423C21547014 @default.
- W3138678423 hasConceptScore W3138678423C41008148 @default.
- W3138678423 hasConceptScore W3138678423C75294576 @default.
- W3138678423 hasConceptScore W3138678423C81363708 @default.
- W3138678423 hasFunder F4320327931 @default.
- W3138678423 hasFunder F4320338453 @default.
- W3138678423 hasLocation W31386784231 @default.
- W3138678423 hasLocation W31386784232 @default.
- W3138678423 hasLocation W31386784233 @default.
- W3138678423 hasLocation W31386784234 @default.
- W3138678423 hasLocation W31386784235 @default.
- W3138678423 hasLocation W31386784236 @default.
- W3138678423 hasOpenAccess W3138678423 @default.
- W3138678423 hasPrimaryLocation W31386784231 @default.
- W3138678423 hasRelatedWork W2214821121 @default.
- W3138678423 hasRelatedWork W2766604260 @default.
- W3138678423 hasRelatedWork W2986507176 @default.
- W3138678423 hasRelatedWork W3012393889 @default.
- W3138678423 hasRelatedWork W3021430260 @default.
- W3138678423 hasRelatedWork W3027997911 @default.
- W3138678423 hasRelatedWork W3144634534 @default.
- W3138678423 hasRelatedWork W3160711233 @default.
- W3138678423 hasRelatedWork W4287776258 @default.
- W3138678423 hasRelatedWork W4384103574 @default.
- W3138678423 hasVolume "4" @default.
- W3138678423 isParatext "false" @default.
- W3138678423 isRetracted "false" @default.
- W3138678423 magId "3138678423" @default.
- W3138678423 workType "article" @default.