Matches in SemOpenAlex for { <https://semopenalex.org/work/W2949750828> ?p ?o ?g. }
Showing items 1 to 84 of
84
with 100 items per page.
- W2949750828 abstract "Methods of transfer learning try to combine knowledge from several related tasks (or domains) to improve performance on a test task. Inspired by causal methodology, we relax the usual covariate shift assumption and assume that it holds true for a subset of predictor variables: the conditional distribution of the target variable given this subset of predictors is invariant over all tasks. We show how this assumption can be motivated from ideas in the field of causality. We focus on the problem of Domain Generalization, in which no examples from the test task are observed. We prove that in an adversarial setting using this subset for prediction is optimal in Domain Generalization; we further provide examples, in which the tasks are sufficiently diverse and the estimator therefore outperforms pooling the data, even on average. If examples from the test task are available, we also provide a method to transfer knowledge from the training tasks and exploit all available features for prediction. However, we provide no guarantees for this method. We introduce a practical method which allows for automatic inference of the above subset and provide corresponding code. We present results on synthetic data sets and a gene deletion data set." @default.
- W2949750828 created "2019-06-27" @default.
- W2949750828 creator A5002127148 @default.
- W2949750828 creator A5044005697 @default.
- W2949750828 creator A5072490106 @default.
- W2949750828 creator A5089280179 @default.
- W2949750828 date "2015-07-19" @default.
- W2949750828 modified "2023-09-27" @default.
- W2949750828 title "Invariant Models for Causal Transfer Learning" @default.
- W2949750828 hasPublicationYear "2015" @default.
- W2949750828 type Work @default.
- W2949750828 sameAs 2949750828 @default.
- W2949750828 citedByCount "1" @default.
- W2949750828 countsByYear W29497508282015 @default.
- W2949750828 crossrefType "posted-content" @default.
- W2949750828 hasAuthorship W2949750828A5002127148 @default.
- W2949750828 hasAuthorship W2949750828A5044005697 @default.
- W2949750828 hasAuthorship W2949750828A5072490106 @default.
- W2949750828 hasAuthorship W2949750828A5089280179 @default.
- W2949750828 hasConcept C105795698 @default.
- W2949750828 hasConcept C119857082 @default.
- W2949750828 hasConcept C134306372 @default.
- W2949750828 hasConcept C149782125 @default.
- W2949750828 hasConcept C150899416 @default.
- W2949750828 hasConcept C154945302 @default.
- W2949750828 hasConcept C158600405 @default.
- W2949750828 hasConcept C165696696 @default.
- W2949750828 hasConcept C177148314 @default.
- W2949750828 hasConcept C185429906 @default.
- W2949750828 hasConcept C190470478 @default.
- W2949750828 hasConcept C2776214188 @default.
- W2949750828 hasConcept C33923547 @default.
- W2949750828 hasConcept C37914503 @default.
- W2949750828 hasConcept C38652104 @default.
- W2949750828 hasConcept C41008148 @default.
- W2949750828 hasConcept C70437156 @default.
- W2949750828 hasConcept C79772020 @default.
- W2949750828 hasConcept C80444323 @default.
- W2949750828 hasConceptScore W2949750828C105795698 @default.
- W2949750828 hasConceptScore W2949750828C119857082 @default.
- W2949750828 hasConceptScore W2949750828C134306372 @default.
- W2949750828 hasConceptScore W2949750828C149782125 @default.
- W2949750828 hasConceptScore W2949750828C150899416 @default.
- W2949750828 hasConceptScore W2949750828C154945302 @default.
- W2949750828 hasConceptScore W2949750828C158600405 @default.
- W2949750828 hasConceptScore W2949750828C165696696 @default.
- W2949750828 hasConceptScore W2949750828C177148314 @default.
- W2949750828 hasConceptScore W2949750828C185429906 @default.
- W2949750828 hasConceptScore W2949750828C190470478 @default.
- W2949750828 hasConceptScore W2949750828C2776214188 @default.
- W2949750828 hasConceptScore W2949750828C33923547 @default.
- W2949750828 hasConceptScore W2949750828C37914503 @default.
- W2949750828 hasConceptScore W2949750828C38652104 @default.
- W2949750828 hasConceptScore W2949750828C41008148 @default.
- W2949750828 hasConceptScore W2949750828C70437156 @default.
- W2949750828 hasConceptScore W2949750828C79772020 @default.
- W2949750828 hasConceptScore W2949750828C80444323 @default.
- W2949750828 hasLocation W29497508281 @default.
- W2949750828 hasOpenAccess W2949750828 @default.
- W2949750828 hasPrimaryLocation W29497508281 @default.
- W2949750828 hasRelatedWork W1960700909 @default.
- W2949750828 hasRelatedWork W2098708428 @default.
- W2949750828 hasRelatedWork W2120708938 @default.
- W2949750828 hasRelatedWork W2126443908 @default.
- W2949750828 hasRelatedWork W2131944683 @default.
- W2949750828 hasRelatedWork W2158760659 @default.
- W2949750828 hasRelatedWork W2604992821 @default.
- W2949750828 hasRelatedWork W2608314995 @default.
- W2949750828 hasRelatedWork W2787551658 @default.
- W2949750828 hasRelatedWork W2798372101 @default.
- W2949750828 hasRelatedWork W2950515451 @default.
- W2949750828 hasRelatedWork W2953096735 @default.
- W2949750828 hasRelatedWork W2964317900 @default.
- W2949750828 hasRelatedWork W3006253006 @default.
- W2949750828 hasRelatedWork W3034466573 @default.
- W2949750828 hasRelatedWork W3092561520 @default.
- W2949750828 hasRelatedWork W3100556893 @default.
- W2949750828 hasRelatedWork W3129086097 @default.
- W2949750828 hasRelatedWork W3202634766 @default.
- W2949750828 hasRelatedWork W2618352547 @default.
- W2949750828 isParatext "false" @default.
- W2949750828 isRetracted "false" @default.
- W2949750828 magId "2949750828" @default.
- W2949750828 workType "article" @default.