Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384157164> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W4384157164 endingPage "110800" @default.
- W4384157164 startingPage "110800" @default.
- W4384157164 abstract "Meta-learning becomes a promising way to solve the few-shot learning problem in recent research. This paradigm mainly relies on hierarchical architecture and episodic training to achieve good generalization on the new learning task. However, as we know, support and query sets in each training task in episodic training only contain a few data points. Too limited training data can lead to data discrepancy between the training and test data which further hurts the performance of learning models in classical supervised and semi-supervised learning. Hence, there is an interesting question: Can we also improve the meta-learning model in few-shot learning by aligning the data distributions of support and query sets? In this work, we try to answer this question. Firstly, we give a theoretical analysis of the generalization ability of meta-learning to show that eliminating the data-distribution discrepancy between the support and query sets in each training task can improve the meta-learning model in few-shot learning. Next, we design an experiment to further verify our analysis. The experimental results show that aligning the data distributions is crucial and even can help the learning model trained by the classical batch-based training achieve better performance on unseen few-shot learning tasks. Furthermore, we also propose two simple meta-batch normalization methods following our result which can be integrated into recent meta-learning methods to improve their performance in few-shot learning." @default.
- W4384157164 created "2023-07-14" @default.
- W4384157164 creator A5005561827 @default.
- W4384157164 creator A5062917240 @default.
- W4384157164 date "2023-10-01" @default.
- W4384157164 modified "2023-09-26" @default.
- W4384157164 title "Can we improve meta-learning model in few-shot learning by aligning data distributions?" @default.
- W4384157164 cites W2104094955 @default.
- W4384157164 cites W2115733720 @default.
- W4384157164 cites W2117539524 @default.
- W4384157164 cites W2165698076 @default.
- W4384157164 cites W2194321275 @default.
- W4384157164 cites W2194775991 @default.
- W4384157164 cites W2257979135 @default.
- W4384157164 cites W2786808285 @default.
- W4384157164 cites W2793888044 @default.
- W4384157164 cites W2963070905 @default.
- W4384157164 cites W2964105864 @default.
- W4384157164 cites W2964983698 @default.
- W4384157164 cites W2981571980 @default.
- W4384157164 cites W3034942609 @default.
- W4384157164 cites W3043239945 @default.
- W4384157164 cites W3163842339 @default.
- W4384157164 cites W4280513799 @default.
- W4384157164 cites W4306975314 @default.
- W4384157164 cites W4317935325 @default.
- W4384157164 doi "https://doi.org/10.1016/j.knosys.2023.110800" @default.
- W4384157164 hasPublicationYear "2023" @default.
- W4384157164 type Work @default.
- W4384157164 citedByCount "0" @default.
- W4384157164 crossrefType "journal-article" @default.
- W4384157164 hasAuthorship W4384157164A5005561827 @default.
- W4384157164 hasAuthorship W4384157164A5062917240 @default.
- W4384157164 hasConcept C115903097 @default.
- W4384157164 hasConcept C119857082 @default.
- W4384157164 hasConcept C134306372 @default.
- W4384157164 hasConcept C136389625 @default.
- W4384157164 hasConcept C136886441 @default.
- W4384157164 hasConcept C144024400 @default.
- W4384157164 hasConcept C145420912 @default.
- W4384157164 hasConcept C154945302 @default.
- W4384157164 hasConcept C162324750 @default.
- W4384157164 hasConcept C177148314 @default.
- W4384157164 hasConcept C187736073 @default.
- W4384157164 hasConcept C19165224 @default.
- W4384157164 hasConcept C24138899 @default.
- W4384157164 hasConcept C2780451532 @default.
- W4384157164 hasConcept C2781002164 @default.
- W4384157164 hasConcept C28006648 @default.
- W4384157164 hasConcept C2986563244 @default.
- W4384157164 hasConcept C33923547 @default.
- W4384157164 hasConcept C41008148 @default.
- W4384157164 hasConcept C50644808 @default.
- W4384157164 hasConcept C58973888 @default.
- W4384157164 hasConceptScore W4384157164C115903097 @default.
- W4384157164 hasConceptScore W4384157164C119857082 @default.
- W4384157164 hasConceptScore W4384157164C134306372 @default.
- W4384157164 hasConceptScore W4384157164C136389625 @default.
- W4384157164 hasConceptScore W4384157164C136886441 @default.
- W4384157164 hasConceptScore W4384157164C144024400 @default.
- W4384157164 hasConceptScore W4384157164C145420912 @default.
- W4384157164 hasConceptScore W4384157164C154945302 @default.
- W4384157164 hasConceptScore W4384157164C162324750 @default.
- W4384157164 hasConceptScore W4384157164C177148314 @default.
- W4384157164 hasConceptScore W4384157164C187736073 @default.
- W4384157164 hasConceptScore W4384157164C19165224 @default.
- W4384157164 hasConceptScore W4384157164C24138899 @default.
- W4384157164 hasConceptScore W4384157164C2780451532 @default.
- W4384157164 hasConceptScore W4384157164C2781002164 @default.
- W4384157164 hasConceptScore W4384157164C28006648 @default.
- W4384157164 hasConceptScore W4384157164C2986563244 @default.
- W4384157164 hasConceptScore W4384157164C33923547 @default.
- W4384157164 hasConceptScore W4384157164C41008148 @default.
- W4384157164 hasConceptScore W4384157164C50644808 @default.
- W4384157164 hasConceptScore W4384157164C58973888 @default.
- W4384157164 hasLocation W43841571641 @default.
- W4384157164 hasOpenAccess W4384157164 @default.
- W4384157164 hasPrimaryLocation W43841571641 @default.
- W4384157164 hasRelatedWork W184546935 @default.
- W4384157164 hasRelatedWork W2091347716 @default.
- W4384157164 hasRelatedWork W2130903752 @default.
- W4384157164 hasRelatedWork W3094076422 @default.
- W4384157164 hasRelatedWork W3169943726 @default.
- W4384157164 hasRelatedWork W3196155444 @default.
- W4384157164 hasRelatedWork W3200361725 @default.
- W4384157164 hasRelatedWork W4285260836 @default.
- W4384157164 hasRelatedWork W4319309271 @default.
- W4384157164 hasRelatedWork W4320063314 @default.
- W4384157164 hasVolume "277" @default.
- W4384157164 isParatext "false" @default.
- W4384157164 isRetracted "false" @default.
- W4384157164 workType "article" @default.