FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312478052> ?p ?o ?g. }

• W4312478052 endingPage "7924" @default.
• W4312478052 startingPage "7914" @default.
• W4312478052 abstract "Data modeling and online monitoring are two critical stages for data-driven anomaly detection. Regarding data modeling, deep neural networks (DNNs) can learn good decision boundaries to separate the anomaly and normal regions, due to their flexible model structures and excellent fitting ability. However, DNNs, using nonlinear activations with specific boundaries, may indirectly cause a limited anomaly detection margin, especially when there are samples far from centroids. Moreover, an anomaly detection model with a narrow detection margin is deemed insensitive to general faults. An anomaly detection model with a tight detection margin will suffer a severe performance degradation. To mitigate the intrinsic drawbacks of DNNs, we develop a new regularizer based on the maximum likelihood of complete data (i.e., observations and latent variables). The regularizer is neuronwise and mathematically acts as compressing neurons, dragging the marginal points into the centroids. Combining the regularizer with the encoding–decoding structure networks, we perform an industrial case study to verify the superiority of the proposed method." @default.
• W4312478052 created "2023-01-05" @default.
• W4312478052 creator A5016865576 @default.
• W4312478052 creator A5018159510 @default.
• W4312478052 creator A5019075429 @default.
• W4312478052 creator A5021184976 @default.
• W4312478052 creator A5046975427 @default.
• W4312478052 creator A5083752100 @default.
• W4312478052 date "2023-07-01" @default.
• W4312478052 modified "2023-09-23" @default.
• W4312478052 title "Neuron-Compressed Deep Neural Network and Its Application in Industrial Anomaly Detection" @default.
• W4312478052 cites W2045979066 @default.
• W4312478052 cites W2060649099 @default.
• W4312478052 cites W2072566913 @default.
• W4312478052 cites W2158204678 @default.
• W4312478052 cites W2158958729 @default.
• W4312478052 cites W2169347809 @default.
• W4312478052 cites W2571164063 @default.
• W4312478052 cites W2741373064 @default.
• W4312478052 cites W2757109865 @default.
• W4312478052 cites W2768047177 @default.
• W4312478052 cites W2769355916 @default.
• W4312478052 cites W2777815485 @default.
• W4312478052 cites W2789634333 @default.
• W4312478052 cites W2794239758 @default.
• W4312478052 cites W2807998075 @default.
• W4312478052 cites W2901910304 @default.
• W4312478052 cites W2916182456 @default.
• W4312478052 cites W2919115771 @default.
• W4312478052 cites W2945532623 @default.
• W4312478052 cites W2946132563 @default.
• W4312478052 cites W2969946296 @default.
• W4312478052 cites W2983926857 @default.
• W4312478052 cites W2984361892 @default.
• W4312478052 cites W2994737449 @default.
• W4312478052 cites W2994908203 @default.
• W4312478052 cites W3001599259 @default.
• W4312478052 cites W3034254119 @default.
• W4312478052 cites W3037099311 @default.
• W4312478052 cites W3081318531 @default.
• W4312478052 cites W3157487677 @default.
• W4312478052 cites W567350711 @default.
• W4312478052 doi "https://doi.org/10.1109/tii.2022.3213819" @default.
• W4312478052 hasPublicationYear "2023" @default.
• W4312478052 type Work @default.
• W4312478052 citedByCount "0" @default.
• W4312478052 crossrefType "journal-article" @default.
• W4312478052 hasAuthorship W4312478052A5016865576 @default.
• W4312478052 hasAuthorship W4312478052A5018159510 @default.
• W4312478052 hasAuthorship W4312478052A5019075429 @default.
• W4312478052 hasAuthorship W4312478052A5021184976 @default.
• W4312478052 hasAuthorship W4312478052A5046975427 @default.
• W4312478052 hasAuthorship W4312478052A5083752100 @default.
• W4312478052 hasConcept C11413529 @default.
• W4312478052 hasConcept C119857082 @default.
• W4312478052 hasConcept C121332964 @default.
• W4312478052 hasConcept C124101348 @default.
• W4312478052 hasConcept C125411270 @default.
• W4312478052 hasConcept C12997251 @default.
• W4312478052 hasConcept C146599234 @default.
• W4312478052 hasConcept C153180895 @default.
• W4312478052 hasConcept C154945302 @default.
• W4312478052 hasConcept C26873012 @default.
• W4312478052 hasConcept C41008148 @default.
• W4312478052 hasConcept C50644808 @default.
• W4312478052 hasConcept C57273362 @default.
• W4312478052 hasConcept C67186912 @default.
• W4312478052 hasConcept C739882 @default.
• W4312478052 hasConcept C77088390 @default.
• W4312478052 hasConcept C774472 @default.
• W4312478052 hasConceptScore W4312478052C11413529 @default.
• W4312478052 hasConceptScore W4312478052C119857082 @default.
• W4312478052 hasConceptScore W4312478052C121332964 @default.
• W4312478052 hasConceptScore W4312478052C124101348 @default.
• W4312478052 hasConceptScore W4312478052C125411270 @default.
• W4312478052 hasConceptScore W4312478052C12997251 @default.
• W4312478052 hasConceptScore W4312478052C146599234 @default.
• W4312478052 hasConceptScore W4312478052C153180895 @default.
• W4312478052 hasConceptScore W4312478052C154945302 @default.
• W4312478052 hasConceptScore W4312478052C26873012 @default.
• W4312478052 hasConceptScore W4312478052C41008148 @default.
• W4312478052 hasConceptScore W4312478052C50644808 @default.
• W4312478052 hasConceptScore W4312478052C57273362 @default.
• W4312478052 hasConceptScore W4312478052C67186912 @default.
• W4312478052 hasConceptScore W4312478052C739882 @default.
• W4312478052 hasConceptScore W4312478052C77088390 @default.
• W4312478052 hasConceptScore W4312478052C774472 @default.
• W4312478052 hasFunder F4320322866 @default.
• W4312478052 hasFunder F4320331088 @default.
• W4312478052 hasIssue "7" @default.
• W4312478052 hasLocation W43124780521 @default.
• W4312478052 hasOpenAccess W4312478052 @default.
• W4312478052 hasPrimaryLocation W43124780521 @default.
• W4312478052 hasRelatedWork W2042251007 @default.
• W4312478052 hasRelatedWork W2065643612 @default.
• W4312478052 hasRelatedWork W2076520961 @default.
• W4312478052 hasRelatedWork W2110365568 @default.
• W4312478052 hasRelatedWork W2966277378 @default.

• next