FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3205743929> ?p ?o ?g. }

• W3205743929 endingPage "3306" @default.
• W3205743929 startingPage "3292" @default.
• W3205743929 abstract "Audio tagging is an active research area and has a wide range of applications. Since the release of AudioSet, great progress has been made in advancing model performance, which mostly comes from the development of novel model architectures and attention modules. However, we find that appropriate training techniques are equally important for building audio tagging models with AudioSet, but have not received the attention they deserve. To fill the gap, in this work, we present PSLA, a collection of training techniques that can noticeably boost the model accuracy including ImageNet pretraining, balanced sampling, data augmentation, label enhancement, model aggregation and their design choices. By training an EfficientNet with these techniques, we obtain a single model (with 13.6M parameters) and an ensemble model that achieve mean average precision (mAP) scores of 0.444 and 0.474 on AudioSet, respectively, outperforming the previous best system of 0.439 with 81M parameters. In addition, our model also achieves a new state-of-the-art mAP of 0.567 on FSD50K." @default.
• W3205743929 created "2021-10-25" @default.
• W3205743929 creator A5008107321 @default.
• W3205743929 creator A5039081803 @default.
• W3205743929 creator A5067285379 @default.
• W3205743929 date "2021-01-01" @default.
• W3205743929 modified "2023-10-04" @default.
• W3205743929 title "PSLA: Improving Audio Tagging With Pretraining, Sampling, Labeling, and Aggregation" @default.
• W3205743929 cites W1556182636 @default.
• W3205743929 cites W1677182931 @default.
• W3205743929 cites W1976526581 @default.
• W3205743929 cites W1995945562 @default.
• W3205743929 cites W2008291900 @default.
• W3205743929 cites W2012296273 @default.
• W3205743929 cites W2038484192 @default.
• W3205743929 cites W2052666245 @default.
• W3205743929 cites W2063125572 @default.
• W3205743929 cites W2097117768 @default.
• W3205743929 cites W2108598243 @default.
• W3205743929 cites W2118978333 @default.
• W3205743929 cites W2155653793 @default.
• W3205743929 cites W2176625348 @default.
• W3205743929 cites W2194775991 @default.
• W3205743929 cites W2593116425 @default.
• W3205743929 cites W2599621350 @default.
• W3205743929 cites W2767754137 @default.
• W3205743929 cites W2807977755 @default.
• W3205743929 cites W2936774411 @default.
• W3205743929 cites W2962910554 @default.
• W3205743929 cites W2963022469 @default.
• W3205743929 cites W2963031676 @default.
• W3205743929 cites W2963163009 @default.
• W3205743929 cites W2963446712 @default.
• W3205743929 cites W2963610932 @default.
• W3205743929 cites W2963679447 @default.
• W3205743929 cites W2963918968 @default.
• W3205743929 cites W2964891022 @default.
• W3205743929 cites W2972943112 @default.
• W3205743929 cites W2973109987 @default.
• W3205743929 cites W2982325051 @default.
• W3205743929 cites W2982339864 @default.
• W3205743929 cites W2982553397 @default.
• W3205743929 cites W2982623348 @default.
• W3205743929 cites W2991391304 @default.
• W3205743929 cites W2995135322 @default.
• W3205743929 cites W3015949486 @default.
• W3205743929 cites W3021910945 @default.
• W3205743929 cites W3094550259 @default.
• W3205743929 cites W3095727342 @default.
• W3205743929 cites W4212883601 @default.
• W3205743929 doi "https://doi.org/10.1109/taslp.2021.3120633" @default.
• W3205743929 hasPublicationYear "2021" @default.
• W3205743929 type Work @default.
• W3205743929 sameAs 3205743929 @default.
• W3205743929 citedByCount "38" @default.
• W3205743929 countsByYear W32057439292022 @default.
• W3205743929 countsByYear W32057439292023 @default.
• W3205743929 crossrefType "journal-article" @default.
• W3205743929 hasAuthorship W3205743929A5008107321 @default.
• W3205743929 hasAuthorship W3205743929A5039081803 @default.
• W3205743929 hasAuthorship W3205743929A5067285379 @default.
• W3205743929 hasBestOaLocation W32057439292 @default.
• W3205743929 hasConcept C106131492 @default.
• W3205743929 hasConcept C119857082 @default.
• W3205743929 hasConcept C124101348 @default.
• W3205743929 hasConcept C140779682 @default.
• W3205743929 hasConcept C154945302 @default.
• W3205743929 hasConcept C159985019 @default.
• W3205743929 hasConcept C192562407 @default.
• W3205743929 hasConcept C204323151 @default.
• W3205743929 hasConcept C31972630 @default.
• W3205743929 hasConcept C41008148 @default.
• W3205743929 hasConcept C51632099 @default.
• W3205743929 hasConceptScore W3205743929C106131492 @default.
• W3205743929 hasConceptScore W3205743929C119857082 @default.
• W3205743929 hasConceptScore W3205743929C124101348 @default.
• W3205743929 hasConceptScore W3205743929C140779682 @default.
• W3205743929 hasConceptScore W3205743929C154945302 @default.
• W3205743929 hasConceptScore W3205743929C159985019 @default.
• W3205743929 hasConceptScore W3205743929C192562407 @default.
• W3205743929 hasConceptScore W3205743929C204323151 @default.
• W3205743929 hasConceptScore W3205743929C31972630 @default.
• W3205743929 hasConceptScore W3205743929C41008148 @default.
• W3205743929 hasConceptScore W3205743929C51632099 @default.
• W3205743929 hasLocation W32057439291 @default.
• W3205743929 hasLocation W32057439292 @default.
• W3205743929 hasOpenAccess W3205743929 @default.
• W3205743929 hasPrimaryLocation W32057439291 @default.
• W3205743929 hasRelatedWork W2961085424 @default.
• W3205743929 hasRelatedWork W3046775127 @default.
• W3205743929 hasRelatedWork W3170094116 @default.
• W3205743929 hasRelatedWork W3209574120 @default.
• W3205743929 hasRelatedWork W4205958290 @default.
• W3205743929 hasRelatedWork W4285260836 @default.
• W3205743929 hasRelatedWork W4286629047 @default.
• W3205743929 hasRelatedWork W4306321456 @default.
• W3205743929 hasRelatedWork W4306674287 @default.
• W3205743929 hasRelatedWork W4224009465 @default.

• next