Matches in SemOpenAlex for { <https://semopenalex.org/work/W2922913792> ?p ?o ?g. }
- W2922913792 abstract "Typical multi-task learning (MTL) methods rely on architectural adjustments and a large trainable parameter set to jointly optimize over several tasks. However, when the number of tasks increases so do the complexity of the architectural adjustments and resource requirements. In this paper, we introduce a method which applies a conditional feature-wise transformation over the convolutional activations that enables a model to successfully perform a large number of tasks. To distinguish from regular MTL, we introduce Many Task Learning (MaTL) as a special case of MTL where more than 20 tasks are performed by a single model. Our method dubbed Task Routing (TR) is encapsulated in a layer we call the Task Routing Layer (TRL), which applied in an MaTL scenario successfully fits hundreds of classification tasks in one model. We evaluate our method on 5 datasets against strong baselines and state-of-the-art approaches." @default.
- W2922913792 created "2019-04-01" @default.
- W2922913792 creator A5070684680 @default.
- W2922913792 creator A5081881122 @default.
- W2922913792 creator A5085795565 @default.
- W2922913792 date "2019-03-28" @default.
- W2922913792 modified "2023-09-28" @default.
- W2922913792 title "Many Task Learning with Task Routing" @default.
- W2922913792 cites W1686810756 @default.
- W2922913792 cites W1797268635 @default.
- W2922913792 cites W1896424170 @default.
- W2922913792 cites W1958932515 @default.
- W2922913792 cites W2078224158 @default.
- W2922913792 cites W2095705004 @default.
- W2922913792 cites W2097451239 @default.
- W2922913792 cites W2119187866 @default.
- W2922913792 cites W2131479143 @default.
- W2922913792 cites W2165698076 @default.
- W2922913792 cites W2194775991 @default.
- W2922913792 cites W2407277018 @default.
- W2922913792 cites W2549401308 @default.
- W2922913792 cites W2562137921 @default.
- W2922913792 cites W2577325523 @default.
- W2922913792 cites W2742079690 @default.
- W2922913792 cites W2750384547 @default.
- W2922913792 cites W2785430118 @default.
- W2922913792 cites W2787189300 @default.
- W2922913792 cites W2791708893 @default.
- W2922913792 cites W2798372101 @default.
- W2922913792 cites W2884886306 @default.
- W2922913792 cites W2886067286 @default.
- W2922913792 cites W2890166761 @default.
- W2922913792 cites W2900964459 @default.
- W2922913792 cites W2903105043 @default.
- W2922913792 cites W2913340405 @default.
- W2922913792 cites W2951478888 @default.
- W2922913792 cites W2951583185 @default.
- W2922913792 cites W2952339051 @default.
- W2922913792 cites W2963072899 @default.
- W2922913792 cites W2963077256 @default.
- W2922913792 cites W2963216850 @default.
- W2922913792 cites W2963258075 @default.
- W2922913792 cites W2963268748 @default.
- W2922913792 cites W2963377935 @default.
- W2922913792 cites W2963446712 @default.
- W2922913792 cites W2963498646 @default.
- W2922913792 cites W2964081403 @default.
- W2922913792 cites W2964185501 @default.
- W2922913792 cites W4919037 @default.
- W2922913792 hasPublicationYear "2019" @default.
- W2922913792 type Work @default.
- W2922913792 sameAs 2922913792 @default.
- W2922913792 citedByCount "2" @default.
- W2922913792 countsByYear W29229137922019 @default.
- W2922913792 countsByYear W29229137922020 @default.
- W2922913792 crossrefType "posted-content" @default.
- W2922913792 hasAuthorship W2922913792A5070684680 @default.
- W2922913792 hasAuthorship W2922913792A5081881122 @default.
- W2922913792 hasAuthorship W2922913792A5085795565 @default.
- W2922913792 hasConcept C104317684 @default.
- W2922913792 hasConcept C119857082 @default.
- W2922913792 hasConcept C138885662 @default.
- W2922913792 hasConcept C154945302 @default.
- W2922913792 hasConcept C162324750 @default.
- W2922913792 hasConcept C177264268 @default.
- W2922913792 hasConcept C178790620 @default.
- W2922913792 hasConcept C185592680 @default.
- W2922913792 hasConcept C187736073 @default.
- W2922913792 hasConcept C199360897 @default.
- W2922913792 hasConcept C204241405 @default.
- W2922913792 hasConcept C2776401178 @default.
- W2922913792 hasConcept C2779227376 @default.
- W2922913792 hasConcept C2780451532 @default.
- W2922913792 hasConcept C28006648 @default.
- W2922913792 hasConcept C31258907 @default.
- W2922913792 hasConcept C41008148 @default.
- W2922913792 hasConcept C41895202 @default.
- W2922913792 hasConcept C55493867 @default.
- W2922913792 hasConcept C74172769 @default.
- W2922913792 hasConceptScore W2922913792C104317684 @default.
- W2922913792 hasConceptScore W2922913792C119857082 @default.
- W2922913792 hasConceptScore W2922913792C138885662 @default.
- W2922913792 hasConceptScore W2922913792C154945302 @default.
- W2922913792 hasConceptScore W2922913792C162324750 @default.
- W2922913792 hasConceptScore W2922913792C177264268 @default.
- W2922913792 hasConceptScore W2922913792C178790620 @default.
- W2922913792 hasConceptScore W2922913792C185592680 @default.
- W2922913792 hasConceptScore W2922913792C187736073 @default.
- W2922913792 hasConceptScore W2922913792C199360897 @default.
- W2922913792 hasConceptScore W2922913792C204241405 @default.
- W2922913792 hasConceptScore W2922913792C2776401178 @default.
- W2922913792 hasConceptScore W2922913792C2779227376 @default.
- W2922913792 hasConceptScore W2922913792C2780451532 @default.
- W2922913792 hasConceptScore W2922913792C28006648 @default.
- W2922913792 hasConceptScore W2922913792C31258907 @default.
- W2922913792 hasConceptScore W2922913792C41008148 @default.
- W2922913792 hasConceptScore W2922913792C41895202 @default.
- W2922913792 hasConceptScore W2922913792C55493867 @default.
- W2922913792 hasConceptScore W2922913792C74172769 @default.
- W2922913792 hasOpenAccess W2922913792 @default.