FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W819164570> ?p ?o ?g. }

• W819164570 abstract "Our main interest is in the problem of making predictions in the online mode of learning where at every step in time a signal arrives and a prediction needs to be made before the corresponding outcome arrives. Loss is suffered if the prediction and outcome do not match perfectly. In the prediction with expert advice framework, this protocol is augmented by a pool of experts that produce their predictions before we have to make ours. The Aggregating Algorithm (AA) is a technique that optimally merges these experts so that the resulting strategy suffers a cumulative loss that is almost as good as that of the best expert in the pool. The AA was applied to the problem of regression, where outcomes are continuous real numbers, to get the AA for Regression (AAR) and its kernel version, KAAR. On typical datasets, KAAR’s empirical performance is not as good as that of Kernel Ridge Regression (KRR) which is a popular regression method. KAAR performs better than KRR only when the data is corrupted with lots of noise or contains severe outliers. To alleviate this we introduce methods that are a hybrid between KRR and KAAR. Empirical experiments suggest that, in general, these new methods perform as good as or better than both KRR and KAAR. In the second part of this dissertation we deal with a more difficult problem — we allow the dependence of outcomes on signals to change with time. To handle this we propose two new methods: WeCKAAR and KAARCh. WeCKAAR is a simple modification of one of our methods from the first part of the dissertation to include decaying weights. KAARCh is an application of the AA to the case where the experts are all the predictors that can change with time. We show that KAARCh suffers a cumulative loss that is almost as good as that of any expert that does not change very rapidly. Empirical results on data with changing dependencies demonstrate that WeCKAAR and KAARCh perform well in practice and are considerably better than Kernel Ridge Regression." @default.
• W819164570 created "2016-06-24" @default.
• W819164570 creator A5003362610 @default.
• W819164570 date "2008-01-01" @default.
• W819164570 modified "2023-09-27" @default.
• W819164570 title "The aggregating algorithm and regression" @default.
• W819164570 cites W1510073064 @default.
• W819164570 cites W1526146785 @default.
• W819164570 cites W1553101044 @default.
• W819164570 cites W1560724230 @default.
• W819164570 cites W1563088657 @default.
• W819164570 cites W1570963478 @default.
• W819164570 cites W1571438983 @default.
• W819164570 cites W1604293137 @default.
• W819164570 cites W1724735345 @default.
• W819164570 cites W1755117326 @default.
• W819164570 cites W1783422035 @default.
• W819164570 cites W1973678258 @default.
• W819164570 cites W1986280275 @default.
• W819164570 cites W1999996900 @default.
• W819164570 cites W2022806724 @default.
• W819164570 cites W2101938294 @default.
• W819164570 cites W2134407776 @default.
• W819164570 cites W2137226992 @default.
• W819164570 cites W2147632820 @default.
• W819164570 cites W2150621701 @default.
• W819164570 cites W2155030296 @default.
• W819164570 cites W2170120409 @default.
• W819164570 cites W2325343629 @default.
• W819164570 cites W2611627047 @default.
• W819164570 cites W2753635250 @default.
• W819164570 cites W2971150588 @default.
• W819164570 cites W3149261725 @default.
• W819164570 cites W772097163 @default.
• W819164570 cites W2139716865 @default.
• W819164570 cites W2498545457 @default.
• W819164570 hasPublicationYear "2008" @default.
• W819164570 type Work @default.
• W819164570 sameAs 819164570 @default.
• W819164570 citedByCount "2" @default.
• W819164570 crossrefType "dissertation" @default.
• W819164570 hasAuthorship W819164570A5003362610 @default.
• W819164570 hasConcept C105795698 @default.
• W819164570 hasConcept C11413529 @default.
• W819164570 hasConcept C114614502 @default.
• W819164570 hasConcept C119857082 @default.
• W819164570 hasConcept C124101348 @default.
• W819164570 hasConcept C144237770 @default.
• W819164570 hasConcept C148220186 @default.
• W819164570 hasConcept C154945302 @default.
• W819164570 hasConcept C200695384 @default.
• W819164570 hasConcept C33923547 @default.
• W819164570 hasConcept C41008148 @default.
• W819164570 hasConcept C74193536 @default.
• W819164570 hasConcept C79337645 @default.
• W819164570 hasConcept C83546350 @default.
• W819164570 hasConceptScore W819164570C105795698 @default.
• W819164570 hasConceptScore W819164570C11413529 @default.
• W819164570 hasConceptScore W819164570C114614502 @default.
• W819164570 hasConceptScore W819164570C119857082 @default.
• W819164570 hasConceptScore W819164570C124101348 @default.
• W819164570 hasConceptScore W819164570C144237770 @default.
• W819164570 hasConceptScore W819164570C148220186 @default.
• W819164570 hasConceptScore W819164570C154945302 @default.
• W819164570 hasConceptScore W819164570C200695384 @default.
• W819164570 hasConceptScore W819164570C33923547 @default.
• W819164570 hasConceptScore W819164570C41008148 @default.
• W819164570 hasConceptScore W819164570C74193536 @default.
• W819164570 hasConceptScore W819164570C79337645 @default.
• W819164570 hasConceptScore W819164570C83546350 @default.
• W819164570 hasLocation W8191645701 @default.
• W819164570 hasOpenAccess W819164570 @default.
• W819164570 hasPrimaryLocation W8191645701 @default.
• W819164570 hasRelatedWork W1511670714 @default.
• W819164570 hasRelatedWork W1638778035 @default.
• W819164570 hasRelatedWork W2137329805 @default.
• W819164570 hasRelatedWork W2149731329 @default.
• W819164570 hasRelatedWork W2162640352 @default.
• W819164570 hasRelatedWork W2183034839 @default.
• W819164570 hasRelatedWork W2187827799 @default.
• W819164570 hasRelatedWork W2211879833 @default.
• W819164570 hasRelatedWork W2282958566 @default.
• W819164570 hasRelatedWork W2401393052 @default.
• W819164570 hasRelatedWork W2403657077 @default.
• W819164570 hasRelatedWork W2520379808 @default.
• W819164570 hasRelatedWork W2920790122 @default.
• W819164570 hasRelatedWork W2924706473 @default.
• W819164570 hasRelatedWork W2945255120 @default.
• W819164570 hasRelatedWork W2949377985 @default.
• W819164570 hasRelatedWork W2955989140 @default.
• W819164570 hasRelatedWork W3124872564 @default.
• W819164570 hasRelatedWork W3135386061 @default.
• W819164570 hasRelatedWork W940094962 @default.
• W819164570 isParatext "false" @default.
• W819164570 isRetracted "false" @default.
• W819164570 magId "819164570" @default.
• W819164570 workType "dissertation" @default.