Matches in SemOpenAlex for { <https://semopenalex.org/work/W2950249966> ?p ?o ?g. }
- W2950249966 abstract "This study introduces a method to design a curriculum for machine-learning to maximize the efficiency during the training process of deep neural networks (DNNs) for speech emotion recognition. Previous studies in other machine-learning problems have shown the benefits of training a classifier following a curriculum where samples are gradually presented in increasing level of difficulty. For speech emotion recognition, the challenge is to establish a natural order of difficulty in the training set to create the curriculum. We address this problem by assuming that ambiguous samples for humans are also ambiguous for computers. Speech samples are often annotated by multiple evaluators to account for differences in emotion perception across individuals. While some sentences with clear emotional content are consistently annotated, sentences with more ambiguous emotional content present important disagreement between individual evaluations. We propose to use the disagreement between evaluators as a measure of difficulty for the classification task. We propose metrics that quantify the inter-evaluation agreement to define the curriculum for regression problems and binary and multi-class classification problems. The experimental results consistently show that relying on a curriculum based on agreement between human judgments leads to statistically significant improvements over baselines trained without a curriculum." @default.
- W2950249966 created "2019-06-27" @default.
- W2950249966 creator A5040793194 @default.
- W2950249966 creator A5041789037 @default.
- W2950249966 date "2018-05-25" @default.
- W2950249966 modified "2023-09-27" @default.
- W2950249966 title "Curriculum Learning for Speech Emotion Recognition from Crowdsourced Labels" @default.
- W2950249966 cites W1261896931 @default.
- W2950249966 cites W141482854 @default.
- W2950249966 cites W1535520578 @default.
- W2950249966 cites W1559582325 @default.
- W2950249966 cites W1581407678 @default.
- W2950249966 cites W1814633089 @default.
- W2950249966 cites W1923034539 @default.
- W2950249966 cites W1976725440 @default.
- W2950249966 cites W2004303440 @default.
- W2950249966 cites W2007347635 @default.
- W2950249966 cites W2010700035 @default.
- W2950249966 cites W2016839396 @default.
- W2950249966 cites W2018297215 @default.
- W2950249966 cites W2042226171 @default.
- W2950249966 cites W2058787788 @default.
- W2950249966 cites W2085662862 @default.
- W2950249966 cites W2097732741 @default.
- W2950249966 cites W2115941714 @default.
- W2950249966 cites W2123864036 @default.
- W2950249966 cites W2125462608 @default.
- W2950249966 cites W2128475742 @default.
- W2950249966 cites W2129794709 @default.
- W2950249966 cites W2137237872 @default.
- W2950249966 cites W2143350951 @default.
- W2950249966 cites W2146334809 @default.
- W2950249966 cites W2151834591 @default.
- W2950249966 cites W2153720647 @default.
- W2950249966 cites W2154780170 @default.
- W2950249966 cites W2156140659 @default.
- W2950249966 cites W2158139315 @default.
- W2950249966 cites W2158943324 @default.
- W2950249966 cites W2170876097 @default.
- W2950249966 cites W2187089797 @default.
- W2950249966 cites W2342475039 @default.
- W2950249966 cites W2401417847 @default.
- W2950249966 cites W2511566646 @default.
- W2950249966 cites W2525412388 @default.
- W2950249966 cites W2527527138 @default.
- W2950249966 cites W2632855849 @default.
- W2950249966 cites W2732479390 @default.
- W2950249966 cites W2742542661 @default.
- W2950249966 cites W2757445263 @default.
- W2950249966 cites W2786657259 @default.
- W2950249966 cites W2964121744 @default.
- W2950249966 cites W2964327384 @default.
- W2950249966 cites W9014458 @default.
- W2950249966 cites W1559937982 @default.
- W2950249966 hasPublicationYear "2018" @default.
- W2950249966 type Work @default.
- W2950249966 sameAs 2950249966 @default.
- W2950249966 citedByCount "0" @default.
- W2950249966 crossrefType "posted-content" @default.
- W2950249966 hasAuthorship W2950249966A5040793194 @default.
- W2950249966 hasAuthorship W2950249966A5041789037 @default.
- W2950249966 hasConcept C119857082 @default.
- W2950249966 hasConcept C12267149 @default.
- W2950249966 hasConcept C154945302 @default.
- W2950249966 hasConcept C15744967 @default.
- W2950249966 hasConcept C162324750 @default.
- W2950249966 hasConcept C169760540 @default.
- W2950249966 hasConcept C177264268 @default.
- W2950249966 hasConcept C187736073 @default.
- W2950249966 hasConcept C19417346 @default.
- W2950249966 hasConcept C199360897 @default.
- W2950249966 hasConcept C204321447 @default.
- W2950249966 hasConcept C26760741 @default.
- W2950249966 hasConcept C2777438025 @default.
- W2950249966 hasConcept C2780451532 @default.
- W2950249966 hasConcept C28490314 @default.
- W2950249966 hasConcept C41008148 @default.
- W2950249966 hasConcept C47177190 @default.
- W2950249966 hasConcept C66905080 @default.
- W2950249966 hasConcept C95623464 @default.
- W2950249966 hasConceptScore W2950249966C119857082 @default.
- W2950249966 hasConceptScore W2950249966C12267149 @default.
- W2950249966 hasConceptScore W2950249966C154945302 @default.
- W2950249966 hasConceptScore W2950249966C15744967 @default.
- W2950249966 hasConceptScore W2950249966C162324750 @default.
- W2950249966 hasConceptScore W2950249966C169760540 @default.
- W2950249966 hasConceptScore W2950249966C177264268 @default.
- W2950249966 hasConceptScore W2950249966C187736073 @default.
- W2950249966 hasConceptScore W2950249966C19417346 @default.
- W2950249966 hasConceptScore W2950249966C199360897 @default.
- W2950249966 hasConceptScore W2950249966C204321447 @default.
- W2950249966 hasConceptScore W2950249966C26760741 @default.
- W2950249966 hasConceptScore W2950249966C2777438025 @default.
- W2950249966 hasConceptScore W2950249966C2780451532 @default.
- W2950249966 hasConceptScore W2950249966C28490314 @default.
- W2950249966 hasConceptScore W2950249966C41008148 @default.
- W2950249966 hasConceptScore W2950249966C47177190 @default.
- W2950249966 hasConceptScore W2950249966C66905080 @default.
- W2950249966 hasConceptScore W2950249966C95623464 @default.
- W2950249966 hasLocation W29502499661 @default.