Matches in SemOpenAlex for { <https://semopenalex.org/work/W3037333676> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W3037333676 endingPage "1539" @default.
- W3037333676 startingPage "1529" @default.
- W3037333676 abstract "In decision-making systems, it is important to have classifiers that have calibrated uncertainties, with an optimisation objective that can be used for automated model selection and training. Gaussian processes (GPs) provide uncertainty estimates and a marginal likelihood objective, but their weak inductive biases lead to inferior accuracy. This has limited their applicability in certain tasks (e.g. image classification). We propose a translation-insensitive convolutional kernel, which relaxes the translation invariance constraint imposed by previous convolutional GPs. We show how we can use the marginal likelihood to learn the degree of insensitivity. We also reformulate GP image-to-image convolutional mappings as multi-output GPs, leading to deep convolutional GPs. We show experimentally that our new kernel improves performance in both single-layer and deep models. We also demonstrate that our fully Bayesian approach improves on dropout-based Bayesian deep learning methods in terms of uncertainty and marginal likelihood estimates." @default.
- W3037333676 created "2020-07-02" @default.
- W3037333676 creator A5023796924 @default.
- W3037333676 creator A5030860008 @default.
- W3037333676 creator A5085437275 @default.
- W3037333676 creator A5090110068 @default.
- W3037333676 date "2020-06-03" @default.
- W3037333676 modified "2023-09-26" @default.
- W3037333676 title "Bayesian Image Classification with Deep Convolutional Gaussian Processes" @default.
- W3037333676 hasPublicationYear "2020" @default.
- W3037333676 type Work @default.
- W3037333676 sameAs 3037333676 @default.
- W3037333676 citedByCount "12" @default.
- W3037333676 countsByYear W30373336762020 @default.
- W3037333676 countsByYear W30373336762021 @default.
- W3037333676 countsByYear W30373336762022 @default.
- W3037333676 crossrefType "proceedings-article" @default.
- W3037333676 hasAuthorship W3037333676A5023796924 @default.
- W3037333676 hasAuthorship W3037333676A5030860008 @default.
- W3037333676 hasAuthorship W3037333676A5085437275 @default.
- W3037333676 hasAuthorship W3037333676A5090110068 @default.
- W3037333676 hasConcept C107673813 @default.
- W3037333676 hasConcept C108583219 @default.
- W3037333676 hasConcept C114614502 @default.
- W3037333676 hasConcept C115961682 @default.
- W3037333676 hasConcept C119857082 @default.
- W3037333676 hasConcept C121332964 @default.
- W3037333676 hasConcept C153180895 @default.
- W3037333676 hasConcept C154945302 @default.
- W3037333676 hasConcept C163716315 @default.
- W3037333676 hasConcept C2776145597 @default.
- W3037333676 hasConcept C33923547 @default.
- W3037333676 hasConcept C41008148 @default.
- W3037333676 hasConcept C61326573 @default.
- W3037333676 hasConcept C62520636 @default.
- W3037333676 hasConcept C74193536 @default.
- W3037333676 hasConcept C75294576 @default.
- W3037333676 hasConcept C81363708 @default.
- W3037333676 hasConcept C95923904 @default.
- W3037333676 hasConceptScore W3037333676C107673813 @default.
- W3037333676 hasConceptScore W3037333676C108583219 @default.
- W3037333676 hasConceptScore W3037333676C114614502 @default.
- W3037333676 hasConceptScore W3037333676C115961682 @default.
- W3037333676 hasConceptScore W3037333676C119857082 @default.
- W3037333676 hasConceptScore W3037333676C121332964 @default.
- W3037333676 hasConceptScore W3037333676C153180895 @default.
- W3037333676 hasConceptScore W3037333676C154945302 @default.
- W3037333676 hasConceptScore W3037333676C163716315 @default.
- W3037333676 hasConceptScore W3037333676C2776145597 @default.
- W3037333676 hasConceptScore W3037333676C33923547 @default.
- W3037333676 hasConceptScore W3037333676C41008148 @default.
- W3037333676 hasConceptScore W3037333676C61326573 @default.
- W3037333676 hasConceptScore W3037333676C62520636 @default.
- W3037333676 hasConceptScore W3037333676C74193536 @default.
- W3037333676 hasConceptScore W3037333676C75294576 @default.
- W3037333676 hasConceptScore W3037333676C81363708 @default.
- W3037333676 hasConceptScore W3037333676C95923904 @default.
- W3037333676 hasLocation W30373336761 @default.
- W3037333676 hasOpenAccess W3037333676 @default.
- W3037333676 hasPrimaryLocation W30373336761 @default.
- W3037333676 hasRelatedWork W137285897 @default.
- W3037333676 hasRelatedWork W1533660737 @default.
- W3037333676 hasRelatedWork W1746819321 @default.
- W3037333676 hasRelatedWork W1777124189 @default.
- W3037333676 hasRelatedWork W2164411961 @default.
- W3037333676 hasRelatedWork W2167608136 @default.
- W3037333676 hasRelatedWork W2809090039 @default.
- W3037333676 hasRelatedWork W2910655610 @default.
- W3037333676 hasRelatedWork W2962685794 @default.
- W3037333676 hasRelatedWork W2962875063 @default.
- W3037333676 hasRelatedWork W2962941879 @default.
- W3037333676 hasRelatedWork W2963323437 @default.
- W3037333676 hasRelatedWork W2963383962 @default.
- W3037333676 hasRelatedWork W2963711523 @default.
- W3037333676 hasRelatedWork W2964052793 @default.
- W3037333676 hasRelatedWork W2964121744 @default.
- W3037333676 hasRelatedWork W3009224979 @default.
- W3037333676 hasRelatedWork W3031355122 @default.
- W3037333676 hasRelatedWork W3037722013 @default.
- W3037333676 hasRelatedWork W3093515944 @default.
- W3037333676 isParatext "false" @default.
- W3037333676 isRetracted "false" @default.
- W3037333676 magId "3037333676" @default.
- W3037333676 workType "article" @default.