FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2802647957> ?p ?o ?g. }

• W2802647957 abstract "Predictive uncertainty is crucial for many computer vision tasks, from image classification to autonomous driving systems. Hamiltonian Monte Carlo (HMC) is an inference method for sampling complex posterior distributions. On the other hand, Dropout regularization has been proposed as an approximate model averaging technique that tends to improve generalization in large scale models such as deep neural networks. Although, HMC provides convergence guarantees for most standard Bayesian models, it does not handle discrete parameters arising from Dropout regularization. In this paper, we present a robust methodology for predictive uncertainty in large scale classification problems, based on Dropout and Stochastic Gradient Hamiltonian Monte Carlo. Even though Dropout induces a non-smooth energy function with no such convergence guarantees, the resulting discretization of the Hamiltonian proves empirical success. The proposed method allows to effectively estimate predictive accuracy and to provide better generalization for difficult test examples." @default.
• W2802647957 created "2018-05-17" @default.
• W2802647957 creator A5013519837 @default.
• W2802647957 creator A5059401123 @default.
• W2802647957 creator A5077127224 @default.
• W2802647957 creator A5084496312 @default.
• W2802647957 date "2018-05-12" @default.
• W2802647957 modified "2023-09-27" @default.
• W2802647957 title "Predictive Uncertainty in Large Scale Classification using Dropout - Stochastic Gradient Hamiltonian Monte Carlo." @default.
• W2802647957 cites W1485147275 @default.
• W2802647957 cites W1506806321 @default.
• W2802647957 cites W1545319692 @default.
• W2802647957 cites W1558572410 @default.
• W2802647957 cites W1585773866 @default.
• W2802647957 cites W1624558029 @default.
• W2802647957 cites W1624701674 @default.
• W2802647957 cites W1981514681 @default.
• W2802647957 cites W2029164135 @default.
• W2802647957 cites W2095705004 @default.
• W2802647957 cites W2112796928 @default.
• W2802647957 cites W2137576513 @default.
• W2802647957 cites W2152799677 @default.
• W2802647957 cites W2159080219 @default.
• W2802647957 cites W2164370980 @default.
• W2802647957 cites W2167433878 @default.
• W2802647957 cites W2231895817 @default.
• W2802647957 cites W2294567968 @default.
• W2802647957 cites W2325939864 @default.
• W2802647957 cites W2473930607 @default.
• W2802647957 cites W2478027467 @default.
• W2802647957 cites W2574963492 @default.
• W2802647957 cites W2577537660 @default.
• W2802647957 cites W2739517340 @default.
• W2802647957 cites W2795604964 @default.
• W2802647957 cites W2951105989 @default.
• W2802647957 cites W2951595529 @default.
• W2802647957 cites W2963977107 @default.
• W2802647957 cites W2964059111 @default.
• W2802647957 cites W2995941077 @default.
• W2802647957 cites W3145164839 @default.
• W2802647957 cites W4919037 @default.
• W2802647957 hasPublicationYear "2018" @default.
• W2802647957 type Work @default.
• W2802647957 sameAs 2802647957 @default.
• W2802647957 citedByCount "0" @default.
• W2802647957 crossrefType "posted-content" @default.
• W2802647957 hasAuthorship W2802647957A5013519837 @default.
• W2802647957 hasAuthorship W2802647957A5059401123 @default.
• W2802647957 hasAuthorship W2802647957A5077127224 @default.
• W2802647957 hasAuthorship W2802647957A5084496312 @default.
• W2802647957 hasConcept C105795698 @default.
• W2802647957 hasConcept C107673813 @default.
• W2802647957 hasConcept C111350023 @default.
• W2802647957 hasConcept C11413529 @default.
• W2802647957 hasConcept C126255220 @default.
• W2802647957 hasConcept C13153151 @default.
• W2802647957 hasConcept C134306372 @default.
• W2802647957 hasConcept C154945302 @default.
• W2802647957 hasConcept C19499675 @default.
• W2802647957 hasConcept C2776135515 @default.
• W2802647957 hasConcept C28826006 @default.
• W2802647957 hasConcept C33923547 @default.
• W2802647957 hasConcept C41008148 @default.
• W2802647957 hasConcept C73000952 @default.
• W2802647957 hasConceptScore W2802647957C105795698 @default.
• W2802647957 hasConceptScore W2802647957C107673813 @default.
• W2802647957 hasConceptScore W2802647957C111350023 @default.
• W2802647957 hasConceptScore W2802647957C11413529 @default.
• W2802647957 hasConceptScore W2802647957C126255220 @default.
• W2802647957 hasConceptScore W2802647957C13153151 @default.
• W2802647957 hasConceptScore W2802647957C134306372 @default.
• W2802647957 hasConceptScore W2802647957C154945302 @default.
• W2802647957 hasConceptScore W2802647957C19499675 @default.
• W2802647957 hasConceptScore W2802647957C2776135515 @default.
• W2802647957 hasConceptScore W2802647957C28826006 @default.
• W2802647957 hasConceptScore W2802647957C33923547 @default.
• W2802647957 hasConceptScore W2802647957C41008148 @default.
• W2802647957 hasConceptScore W2802647957C73000952 @default.
• W2802647957 hasLocation W28026479571 @default.
• W2802647957 hasOpenAccess W2802647957 @default.
• W2802647957 hasPrimaryLocation W28026479571 @default.
• W2802647957 hasRelatedWork W1574102181 @default.
• W2802647957 hasRelatedWork W1678034755 @default.
• W2802647957 hasRelatedWork W1711953686 @default.
• W2802647957 hasRelatedWork W1956541315 @default.
• W2802647957 hasRelatedWork W1972472593 @default.
• W2802647957 hasRelatedWork W2012823783 @default.
• W2802647957 hasRelatedWork W2036405337 @default.
• W2802647957 hasRelatedWork W2041185912 @default.
• W2802647957 hasRelatedWork W2162538033 @default.
• W2802647957 hasRelatedWork W2165609874 @default.
• W2802647957 hasRelatedWork W2202617390 @default.
• W2802647957 hasRelatedWork W2219117298 @default.
• W2802647957 hasRelatedWork W2578632808 @default.
• W2802647957 hasRelatedWork W2893924688 @default.
• W2802647957 hasRelatedWork W2952028434 @default.
• W2802647957 hasRelatedWork W2963217000 @default.
• W2802647957 hasRelatedWork W3038014470 @default.
• W2802647957 hasRelatedWork W3048852403 @default.
• W2802647957 hasRelatedWork W3126797316 @default.

• next