Matches in SemOpenAlex for { <https://semopenalex.org/work/W4297839343> ?p ?o ?g. }
Showing items 1 to 61 of
61
with 100 items per page.
- W4297839343 abstract "Sound event detection (SED) methods are tasked with labeling segments of audio recordings by the presence of active sound sources. SED is typically posed as a supervised machine learning problem, requiring strong annotations for the presence or absence of each sound source at every time instant within the recording. However, strong annotations of this type are both labor- and cost-intensive for human annotators to produce, which limits the practical scalability of SED methods. In this work, we treat SED as a multiple instance learning (MIL) problem, where training labels are static over a short excerpt, indicating the presence or absence of sound sources but not their temporal locality. The models, however, must still produce temporally dynamic predictions, which must be aggregated (pooled) when comparing against static labels during training. To facilitate this aggregation, we develop a family of adaptive pooling operators---referred to as auto-pool---which smoothly interpolate between common pooling operators, such as min-, max-, or average-pooling, and automatically adapt to the characteristics of the sound sources in question. We evaluate the proposed pooling operators on three datasets, and demonstrate that in each case, the proposed methods outperform non-adaptive pooling operators for static prediction, and nearly match the performance of models trained with strong, dynamic annotations. The proposed method is evaluated in conjunction with convolutional neural networks, but can be readily applied to any differentiable model for time-series label prediction." @default.
- W4297839343 created "2022-10-01" @default.
- W4297839343 creator A5010404092 @default.
- W4297839343 creator A5031398497 @default.
- W4297839343 creator A5037548450 @default.
- W4297839343 date "2018-04-26" @default.
- W4297839343 modified "2023-09-26" @default.
- W4297839343 title "Adaptive pooling operators for weakly labeled sound event detection" @default.
- W4297839343 doi "https://doi.org/10.48550/arxiv.1804.10070" @default.
- W4297839343 hasPublicationYear "2018" @default.
- W4297839343 type Work @default.
- W4297839343 citedByCount "0" @default.
- W4297839343 crossrefType "posted-content" @default.
- W4297839343 hasAuthorship W4297839343A5010404092 @default.
- W4297839343 hasAuthorship W4297839343A5031398497 @default.
- W4297839343 hasAuthorship W4297839343A5037548450 @default.
- W4297839343 hasBestOaLocation W42978393431 @default.
- W4297839343 hasConcept C119857082 @default.
- W4297839343 hasConcept C121332964 @default.
- W4297839343 hasConcept C138885662 @default.
- W4297839343 hasConcept C153180895 @default.
- W4297839343 hasConcept C154945302 @default.
- W4297839343 hasConcept C2779662365 @default.
- W4297839343 hasConcept C2779808786 @default.
- W4297839343 hasConcept C41008148 @default.
- W4297839343 hasConcept C41895202 @default.
- W4297839343 hasConcept C48044578 @default.
- W4297839343 hasConcept C62520636 @default.
- W4297839343 hasConcept C70437156 @default.
- W4297839343 hasConcept C77088390 @default.
- W4297839343 hasConcept C81363708 @default.
- W4297839343 hasConceptScore W4297839343C119857082 @default.
- W4297839343 hasConceptScore W4297839343C121332964 @default.
- W4297839343 hasConceptScore W4297839343C138885662 @default.
- W4297839343 hasConceptScore W4297839343C153180895 @default.
- W4297839343 hasConceptScore W4297839343C154945302 @default.
- W4297839343 hasConceptScore W4297839343C2779662365 @default.
- W4297839343 hasConceptScore W4297839343C2779808786 @default.
- W4297839343 hasConceptScore W4297839343C41008148 @default.
- W4297839343 hasConceptScore W4297839343C41895202 @default.
- W4297839343 hasConceptScore W4297839343C48044578 @default.
- W4297839343 hasConceptScore W4297839343C62520636 @default.
- W4297839343 hasConceptScore W4297839343C70437156 @default.
- W4297839343 hasConceptScore W4297839343C77088390 @default.
- W4297839343 hasConceptScore W4297839343C81363708 @default.
- W4297839343 hasLocation W42978393431 @default.
- W4297839343 hasOpenAccess W4297839343 @default.
- W4297839343 hasPrimaryLocation W42978393431 @default.
- W4297839343 hasRelatedWork W2291847203 @default.
- W4297839343 hasRelatedWork W2368694199 @default.
- W4297839343 hasRelatedWork W2424871898 @default.
- W4297839343 hasRelatedWork W2514274290 @default.
- W4297839343 hasRelatedWork W2517027266 @default.
- W4297839343 hasRelatedWork W2613736958 @default.
- W4297839343 hasRelatedWork W2758063741 @default.
- W4297839343 hasRelatedWork W2792080776 @default.
- W4297839343 hasRelatedWork W2940661641 @default.
- W4297839343 hasRelatedWork W2969680539 @default.
- W4297839343 isParatext "false" @default.
- W4297839343 isRetracted "false" @default.
- W4297839343 workType "article" @default.