Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312908045> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W4312908045 endingPage "238" @default.
- W4312908045 startingPage "221" @default.
- W4312908045 abstract "The paradigm of machine intelligence moves from purely supervised learning to a more practical scenario when many loosely related unlabeled data are available and labeled data is scarce. Most existing algorithms assume that the underlying task distribution is stationary. Here we consider a more realistic and challenging setting in that task distributions evolve over time. We name this problem as Semi-supervised meta-learning with Evolving Task diStributions, abbreviated as SETS. Two key challenges arise in this more realistic setting: (i) how to use unlabeled data in the presence of a large amount of unlabeled out-of-distribution (OOD) data; and (ii) how to prevent catastrophic forgetting on previously learned task distributions due to the task distribution shift. We propose an OOD Robust and knowleDge presErved semi-supeRvised meta-learning approach (ORDER) (we use ORDER to denote the task distributions sequentially arrive with some ORDER), to tackle these two major challenges. Specifically, our ORDER introduces a novel mutual information regularization to robustify the model with unlabeled OOD data and adopts an optimal transport regularization to remember previously learned knowledge in feature space. In addition, we test our method on a very challenging dataset: SETS on large-scale non-stationary semi-supervised task distributions consisting of (at least) 72K tasks. With extensive experiments, we demonstrate the proposed ORDER alleviates forgetting on evolving task distributions and is more robust to OOD data than related strong baselines." @default.
- W4312908045 created "2023-01-05" @default.
- W4312908045 creator A5000950850 @default.
- W4312908045 creator A5002091761 @default.
- W4312908045 creator A5024023588 @default.
- W4312908045 creator A5033311476 @default.
- W4312908045 creator A5039291107 @default.
- W4312908045 creator A5044277083 @default.
- W4312908045 creator A5069086470 @default.
- W4312908045 date "2022-01-01" @default.
- W4312908045 modified "2023-10-16" @default.
- W4312908045 title "Meta-Learning with Less Forgetting on Large-Scale Non-Stationary Task Distributions" @default.
- W4312908045 cites W1932198206 @default.
- W4312908045 cites W2119885577 @default.
- W4312908045 cites W2533598788 @default.
- W4312908045 cites W2560647685 @default.
- W4312908045 cites W2885613075 @default.
- W4312908045 cites W2896652510 @default.
- W4312908045 cites W2963070905 @default.
- W4312908045 cites W2963588172 @default.
- W4312908045 cites W2981864462 @default.
- W4312908045 cites W2982220706 @default.
- W4312908045 cites W2982701845 @default.
- W4312908045 cites W3009213340 @default.
- W4312908045 cites W3034930293 @default.
- W4312908045 cites W3035067471 @default.
- W4312908045 cites W3108827478 @default.
- W4312908045 cites W3180003570 @default.
- W4312908045 cites W3181161034 @default.
- W4312908045 cites W3203131677 @default.
- W4312908045 cites W3204328054 @default.
- W4312908045 cites W4312585885 @default.
- W4312908045 doi "https://doi.org/10.1007/978-3-031-20044-1_13" @default.
- W4312908045 hasPublicationYear "2022" @default.
- W4312908045 type Work @default.
- W4312908045 citedByCount "1" @default.
- W4312908045 countsByYear W43129080452023 @default.
- W4312908045 crossrefType "book-chapter" @default.
- W4312908045 hasAuthorship W4312908045A5000950850 @default.
- W4312908045 hasAuthorship W4312908045A5002091761 @default.
- W4312908045 hasAuthorship W4312908045A5024023588 @default.
- W4312908045 hasAuthorship W4312908045A5033311476 @default.
- W4312908045 hasAuthorship W4312908045A5039291107 @default.
- W4312908045 hasAuthorship W4312908045A5044277083 @default.
- W4312908045 hasAuthorship W4312908045A5069086470 @default.
- W4312908045 hasBestOaLocation W43129080452 @default.
- W4312908045 hasConcept C119857082 @default.
- W4312908045 hasConcept C136389625 @default.
- W4312908045 hasConcept C138885662 @default.
- W4312908045 hasConcept C154945302 @default.
- W4312908045 hasConcept C162324750 @default.
- W4312908045 hasConcept C187736073 @default.
- W4312908045 hasConcept C2776135515 @default.
- W4312908045 hasConcept C2776145971 @default.
- W4312908045 hasConcept C2780451532 @default.
- W4312908045 hasConcept C28006648 @default.
- W4312908045 hasConcept C41008148 @default.
- W4312908045 hasConcept C41895202 @default.
- W4312908045 hasConcept C50644808 @default.
- W4312908045 hasConcept C7149132 @default.
- W4312908045 hasConceptScore W4312908045C119857082 @default.
- W4312908045 hasConceptScore W4312908045C136389625 @default.
- W4312908045 hasConceptScore W4312908045C138885662 @default.
- W4312908045 hasConceptScore W4312908045C154945302 @default.
- W4312908045 hasConceptScore W4312908045C162324750 @default.
- W4312908045 hasConceptScore W4312908045C187736073 @default.
- W4312908045 hasConceptScore W4312908045C2776135515 @default.
- W4312908045 hasConceptScore W4312908045C2776145971 @default.
- W4312908045 hasConceptScore W4312908045C2780451532 @default.
- W4312908045 hasConceptScore W4312908045C28006648 @default.
- W4312908045 hasConceptScore W4312908045C41008148 @default.
- W4312908045 hasConceptScore W4312908045C41895202 @default.
- W4312908045 hasConceptScore W4312908045C50644808 @default.
- W4312908045 hasConceptScore W4312908045C7149132 @default.
- W4312908045 hasLocation W43129080451 @default.
- W4312908045 hasLocation W43129080452 @default.
- W4312908045 hasOpenAccess W4312908045 @default.
- W4312908045 hasPrimaryLocation W43129080451 @default.
- W4312908045 hasRelatedWork W2538661024 @default.
- W4312908045 hasRelatedWork W2983785000 @default.
- W4312908045 hasRelatedWork W3023831224 @default.
- W4312908045 hasRelatedWork W3097966281 @default.
- W4312908045 hasRelatedWork W3199131761 @default.
- W4312908045 hasRelatedWork W3216395362 @default.
- W4312908045 hasRelatedWork W4221153249 @default.
- W4312908045 hasRelatedWork W4287781552 @default.
- W4312908045 hasRelatedWork W4319309271 @default.
- W4312908045 hasRelatedWork W4382317652 @default.
- W4312908045 isParatext "false" @default.
- W4312908045 isRetracted "false" @default.
- W4312908045 workType "book-chapter" @default.