Matches in SemOpenAlex for { <https://semopenalex.org/work/W3004679435> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W3004679435 endingPage "28" @default.
- W3004679435 startingPage "1" @default.
- W3004679435 abstract "We examine a new form of smooth approximation to the zero one loss in which learning is performed using a reformulation of the widely used logistic function. Our approach is based on using the posterior mean of a novel generalized Beta-Bernoulli formulation. This leads to a generalized logistic function that approximates the zero one loss, but retains a probabilistic formulation conferring a number of useful properties. The approach is easily generalized to kernel logistic regression and easily integrated into methods for structured prediction. We present experiments in which we learn such models using an optimization method consisting of a combination of gradient descent and coordinate descent using localized grid search so as to escape from local minima. Our experiments indicate that optimization quality is improved when learning metaparameters are themselves optimized using a validation set. Our experiments show improved performance relative to widely used logistic and hinge loss methods on a wide variety of problems ranging from standard UC Irvine and libSVM evaluation datasets to product review predictions and a visual information extraction task. We observe that the approach is as follows: (1) more robust to outliers compared to the logistic and hinge losses; (2) outperforms comparable logistic and max margin models on larger scale benchmark problems; (3) when combined with Gaussian–Laplacian mixture prior on parameters the kernelized version of our formulation yields sparser solutions than Support Vector Machine classifiers; and (4) when integrated into a probabilistic structured prediction technique our approach provides more accurate probabilities yielding improved inference and increasing information extraction performance." @default.
- W3004679435 created "2020-02-14" @default.
- W3004679435 creator A5027084430 @default.
- W3004679435 creator A5075885606 @default.
- W3004679435 date "2019-12-13" @default.
- W3004679435 modified "2023-09-22" @default.
- W3004679435 title "A New Smooth Approximation to the Zero One Loss with a Probabilistic Interpretation" @default.
- W3004679435 cites W1514928307 @default.
- W3004679435 cites W1524688041 @default.
- W3004679435 cites W1663973292 @default.
- W3004679435 cites W1989494851 @default.
- W3004679435 cites W1994720006 @default.
- W3004679435 cites W2014374971 @default.
- W3004679435 cites W2030843733 @default.
- W3004679435 cites W2038663599 @default.
- W3004679435 cites W2068221105 @default.
- W3004679435 cites W2068696370 @default.
- W3004679435 cites W2089394015 @default.
- W3004679435 cites W2122825543 @default.
- W3004679435 cites W2129809168 @default.
- W3004679435 cites W2146666958 @default.
- W3004679435 cites W2162152253 @default.
- W3004679435 cites W2163608767 @default.
- W3004679435 cites W2166765763 @default.
- W3004679435 cites W2469023256 @default.
- W3004679435 cites W2534029181 @default.
- W3004679435 cites W2757686304 @default.
- W3004679435 cites W2761674588 @default.
- W3004679435 cites W3003518101 @default.
- W3004679435 cites W4240374930 @default.
- W3004679435 doi "https://doi.org/10.1145/3365672" @default.
- W3004679435 hasPublicationYear "2019" @default.
- W3004679435 type Work @default.
- W3004679435 sameAs 3004679435 @default.
- W3004679435 citedByCount "0" @default.
- W3004679435 crossrefType "journal-article" @default.
- W3004679435 hasAuthorship W3004679435A5027084430 @default.
- W3004679435 hasAuthorship W3004679435A5075885606 @default.
- W3004679435 hasBestOaLocation W30046794352 @default.
- W3004679435 hasConcept C11413529 @default.
- W3004679435 hasConcept C119857082 @default.
- W3004679435 hasConcept C12267149 @default.
- W3004679435 hasConcept C126255220 @default.
- W3004679435 hasConcept C134306372 @default.
- W3004679435 hasConcept C153180895 @default.
- W3004679435 hasConcept C154945302 @default.
- W3004679435 hasConcept C186633575 @default.
- W3004679435 hasConcept C33923547 @default.
- W3004679435 hasConcept C39891107 @default.
- W3004679435 hasConcept C41008148 @default.
- W3004679435 hasConcept C49937458 @default.
- W3004679435 hasConceptScore W3004679435C11413529 @default.
- W3004679435 hasConceptScore W3004679435C119857082 @default.
- W3004679435 hasConceptScore W3004679435C12267149 @default.
- W3004679435 hasConceptScore W3004679435C126255220 @default.
- W3004679435 hasConceptScore W3004679435C134306372 @default.
- W3004679435 hasConceptScore W3004679435C153180895 @default.
- W3004679435 hasConceptScore W3004679435C154945302 @default.
- W3004679435 hasConceptScore W3004679435C186633575 @default.
- W3004679435 hasConceptScore W3004679435C33923547 @default.
- W3004679435 hasConceptScore W3004679435C39891107 @default.
- W3004679435 hasConceptScore W3004679435C41008148 @default.
- W3004679435 hasConceptScore W3004679435C49937458 @default.
- W3004679435 hasIssue "1" @default.
- W3004679435 hasLocation W30046794351 @default.
- W3004679435 hasLocation W30046794352 @default.
- W3004679435 hasOpenAccess W3004679435 @default.
- W3004679435 hasPrimaryLocation W30046794351 @default.
- W3004679435 hasRelatedWork W2041399278 @default.
- W3004679435 hasRelatedWork W2099369243 @default.
- W3004679435 hasRelatedWork W2120008580 @default.
- W3004679435 hasRelatedWork W2136184105 @default.
- W3004679435 hasRelatedWork W2163073107 @default.
- W3004679435 hasRelatedWork W3194539120 @default.
- W3004679435 hasRelatedWork W4205958290 @default.
- W3004679435 hasRelatedWork W4223656335 @default.
- W3004679435 hasRelatedWork W2187500075 @default.
- W3004679435 hasRelatedWork W2345184372 @default.
- W3004679435 hasVolume "14" @default.
- W3004679435 isParatext "false" @default.
- W3004679435 isRetracted "false" @default.
- W3004679435 magId "3004679435" @default.
- W3004679435 workType "article" @default.