Matches in SemOpenAlex for { <https://semopenalex.org/work/W2405504530> ?p ?o ?g. }
Showing items 1 to 70 of
70
with 100 items per page.
- W2405504530 abstract "We introduce a new approach for designing computationally efficient learning algorithms that are tolerant to noise, and demonstrate its effectiveness by designing algorithms with improved noise tolerance guarantees for learning linear separators. We consider both the malicious noise model and the adversarial label noise model. For malicious noise, where the adversary can corrupt both the label and the features, we provide a polynomial-time algorithm for learning linear separators in $Re^d$ under isotropic log-concave distributions that can tolerate a nearly information-theoretically optimal noise rate of $eta = Omega(epsilon)$. For the adversarial label noise model, where the distribution over the feature vectors is unchanged, and the overall probability of a noisy label is constrained to be at most $eta$, we also give a polynomial-time algorithm for learning linear separators in $Re^d$ under isotropic log-concave distributions that can handle a noise rate of $eta = Omegaleft(epsilonright)$. We show that, in the active learning model, our algorithms achieve a label complexity whose dependence on the error parameter $epsilon$ is polylogarithmic. This provides the first polynomial-time active learning algorithm for learning linear separators in the presence of malicious noise or adversarial label noise." @default.
- W2405504530 created "2016-06-24" @default.
- W2405504530 creator A5056617357 @default.
- W2405504530 creator A5068544954 @default.
- W2405504530 creator A5084487318 @default.
- W2405504530 date "2013-07-31" @default.
- W2405504530 modified "2023-09-26" @default.
- W2405504530 title "The Power of Localization for Efficiently Learning Linear Separators with Malicious Noise." @default.
- W2405504530 hasPublicationYear "2013" @default.
- W2405504530 type Work @default.
- W2405504530 sameAs 2405504530 @default.
- W2405504530 citedByCount "1" @default.
- W2405504530 countsByYear W24055045302013 @default.
- W2405504530 crossrefType "posted-content" @default.
- W2405504530 hasAuthorship W2405504530A5056617357 @default.
- W2405504530 hasAuthorship W2405504530A5068544954 @default.
- W2405504530 hasAuthorship W2405504530A5084487318 @default.
- W2405504530 hasConcept C11413529 @default.
- W2405504530 hasConcept C115961682 @default.
- W2405504530 hasConcept C121332964 @default.
- W2405504530 hasConcept C134306372 @default.
- W2405504530 hasConcept C154945302 @default.
- W2405504530 hasConcept C163258240 @default.
- W2405504530 hasConcept C203234222 @default.
- W2405504530 hasConcept C311688 @default.
- W2405504530 hasConcept C33923547 @default.
- W2405504530 hasConcept C41008148 @default.
- W2405504530 hasConcept C62520636 @default.
- W2405504530 hasConcept C90119067 @default.
- W2405504530 hasConcept C99498987 @default.
- W2405504530 hasConceptScore W2405504530C11413529 @default.
- W2405504530 hasConceptScore W2405504530C115961682 @default.
- W2405504530 hasConceptScore W2405504530C121332964 @default.
- W2405504530 hasConceptScore W2405504530C134306372 @default.
- W2405504530 hasConceptScore W2405504530C154945302 @default.
- W2405504530 hasConceptScore W2405504530C163258240 @default.
- W2405504530 hasConceptScore W2405504530C203234222 @default.
- W2405504530 hasConceptScore W2405504530C311688 @default.
- W2405504530 hasConceptScore W2405504530C33923547 @default.
- W2405504530 hasConceptScore W2405504530C41008148 @default.
- W2405504530 hasConceptScore W2405504530C62520636 @default.
- W2405504530 hasConceptScore W2405504530C90119067 @default.
- W2405504530 hasConceptScore W2405504530C99498987 @default.
- W2405504530 hasLocation W24055045301 @default.
- W2405504530 hasOpenAccess W2405504530 @default.
- W2405504530 hasPrimaryLocation W24055045301 @default.
- W2405504530 hasRelatedWork W100695655 @default.
- W2405504530 hasRelatedWork W1515737631 @default.
- W2405504530 hasRelatedWork W2111895931 @default.
- W2405504530 hasRelatedWork W2166581276 @default.
- W2405504530 hasRelatedWork W2173497155 @default.
- W2405504530 hasRelatedWork W2340619382 @default.
- W2405504530 hasRelatedWork W2475016247 @default.
- W2405504530 hasRelatedWork W2592278891 @default.
- W2405504530 hasRelatedWork W2617991662 @default.
- W2405504530 hasRelatedWork W2795394396 @default.
- W2405504530 hasRelatedWork W2885472922 @default.
- W2405504530 hasRelatedWork W2890489840 @default.
- W2405504530 hasRelatedWork W2951563626 @default.
- W2405504530 hasRelatedWork W2952610888 @default.
- W2405504530 hasRelatedWork W2953340467 @default.
- W2405504530 hasRelatedWork W2968223339 @default.
- W2405504530 hasRelatedWork W3004722026 @default.
- W2405504530 hasRelatedWork W3046885009 @default.
- W2405504530 hasRelatedWork W3123553160 @default.
- W2405504530 hasRelatedWork W3164295722 @default.
- W2405504530 isParatext "false" @default.
- W2405504530 isRetracted "false" @default.
- W2405504530 magId "2405504530" @default.
- W2405504530 workType "article" @default.