Matches in SemOpenAlex for { <https://semopenalex.org/work/W2171566706> ?p ?o ?g. }
- W2171566706 endingPage "2666" @default.
- W2171566706 startingPage "2643" @default.
- W2171566706 abstract "A common problem of kernel-based online algorithms, such as the kernel-based Perceptron algorithm, is the amount of memory required to store the online hypothesis, which may increase without bound as the algorithm progresses. Furthermore, the computational load of such algorithms grows linearly with the amount of memory used to store the hypothesis. To attack these problems, most previous work has focused on discarding some of the instances, in order to keep the memory bounded. In this paper we present a new algorithm, in which the instances are not discarded, but are instead projected onto the space spanned by the previous online hypothesis. We call this algorithm Projectron. While the memory size of the Projectron solution cannot be predicted before training, we prove that its solution is guaranteed to be bounded. We derive a relative mistake bound for the proposed algorithm, and deduce from it a slightly different algorithm which outperforms the Perceptron. We call this second algorithm Projectron++. We show that this algorithm can be extended to handle the multiclass and the structured output settings, resulting, as far as we know, in the first online bounded algorithm that can learn complex classification tasks. The method of bounding the hypothesis representation can be applied to any conservative online algorithm and to other online algorithms, as it is demonstrated for ALMA2. Experimental results on various data sets show the empirical advantage of our technique compared to various bounded online algorithms, both in terms of memory and accuracy." @default.
- W2171566706 created "2016-06-24" @default.
- W2171566706 creator A5008847407 @default.
- W2171566706 creator A5024757596 @default.
- W2171566706 creator A5044260265 @default.
- W2171566706 date "2009-12-01" @default.
- W2171566706 modified "2023-10-10" @default.
- W2171566706 title "Bounded Kernel-Based Online Learning" @default.
- W2171566706 cites W1512098439 @default.
- W2171566706 cites W1540155273 @default.
- W2171566706 cites W1540586255 @default.
- W2171566706 cites W1714704734 @default.
- W2171566706 cites W1762008180 @default.
- W2171566706 cites W1890144016 @default.
- W2171566706 cites W1975588358 @default.
- W2171566706 cites W1979711143 @default.
- W2171566706 cites W2012715465 @default.
- W2171566706 cites W2020467629 @default.
- W2171566706 cites W2040870580 @default.
- W2171566706 cites W2100659887 @default.
- W2171566706 cites W2105644991 @default.
- W2171566706 cites W2107275319 @default.
- W2171566706 cites W2108807072 @default.
- W2171566706 cites W2109339818 @default.
- W2171566706 cites W2112796928 @default.
- W2171566706 cites W2121423746 @default.
- W2171566706 cites W2129564505 @default.
- W2171566706 cites W2132882064 @default.
- W2171566706 cites W2139432235 @default.
- W2171566706 cites W2150621701 @default.
- W2171566706 cites W2153290280 @default.
- W2171566706 cites W2156135142 @default.
- W2171566706 cites W2157017541 @default.
- W2171566706 cites W2160218441 @default.
- W2171566706 cites W2404666544 @default.
- W2171566706 cites W2429914308 @default.
- W2171566706 cites W3021452258 @default.
- W2171566706 cites W597932189 @default.
- W2171566706 cites W93885277 @default.
- W2171566706 hasPublicationYear "2009" @default.
- W2171566706 type Work @default.
- W2171566706 sameAs 2171566706 @default.
- W2171566706 citedByCount "38" @default.
- W2171566706 countsByYear W21715667062012 @default.
- W2171566706 countsByYear W21715667062013 @default.
- W2171566706 countsByYear W21715667062014 @default.
- W2171566706 countsByYear W21715667062015 @default.
- W2171566706 countsByYear W21715667062016 @default.
- W2171566706 countsByYear W21715667062017 @default.
- W2171566706 countsByYear W21715667062018 @default.
- W2171566706 countsByYear W21715667062019 @default.
- W2171566706 countsByYear W21715667062020 @default.
- W2171566706 crossrefType "journal-article" @default.
- W2171566706 hasAuthorship W2171566706A5008847407 @default.
- W2171566706 hasAuthorship W2171566706A5024757596 @default.
- W2171566706 hasAuthorship W2171566706A5044260265 @default.
- W2171566706 hasConcept C11413529 @default.
- W2171566706 hasConcept C118615104 @default.
- W2171566706 hasConcept C119857082 @default.
- W2171566706 hasConcept C134306372 @default.
- W2171566706 hasConcept C154945302 @default.
- W2171566706 hasConcept C196921405 @default.
- W2171566706 hasConcept C33923547 @default.
- W2171566706 hasConcept C34388435 @default.
- W2171566706 hasConcept C41008148 @default.
- W2171566706 hasConcept C50644808 @default.
- W2171566706 hasConcept C60908668 @default.
- W2171566706 hasConcept C63584917 @default.
- W2171566706 hasConcept C74193536 @default.
- W2171566706 hasConcept C77553402 @default.
- W2171566706 hasConceptScore W2171566706C11413529 @default.
- W2171566706 hasConceptScore W2171566706C118615104 @default.
- W2171566706 hasConceptScore W2171566706C119857082 @default.
- W2171566706 hasConceptScore W2171566706C134306372 @default.
- W2171566706 hasConceptScore W2171566706C154945302 @default.
- W2171566706 hasConceptScore W2171566706C196921405 @default.
- W2171566706 hasConceptScore W2171566706C33923547 @default.
- W2171566706 hasConceptScore W2171566706C34388435 @default.
- W2171566706 hasConceptScore W2171566706C41008148 @default.
- W2171566706 hasConceptScore W2171566706C50644808 @default.
- W2171566706 hasConceptScore W2171566706C60908668 @default.
- W2171566706 hasConceptScore W2171566706C63584917 @default.
- W2171566706 hasConceptScore W2171566706C74193536 @default.
- W2171566706 hasConceptScore W2171566706C77553402 @default.
- W2171566706 hasIssue "92" @default.
- W2171566706 hasLocation W21715667061 @default.
- W2171566706 hasOpenAccess W2171566706 @default.
- W2171566706 hasPrimaryLocation W21715667061 @default.
- W2171566706 hasRelatedWork W1540155273 @default.
- W2171566706 hasRelatedWork W1560724230 @default.
- W2171566706 hasRelatedWork W1923046654 @default.
- W2171566706 hasRelatedWork W1973678258 @default.
- W2171566706 hasRelatedWork W1975588358 @default.
- W2171566706 hasRelatedWork W1979711143 @default.
- W2171566706 hasRelatedWork W2040870580 @default.
- W2171566706 hasRelatedWork W2095895508 @default.
- W2171566706 hasRelatedWork W2097645432 @default.
- W2171566706 hasRelatedWork W2108807072 @default.