Matches in SemOpenAlex for { <https://semopenalex.org/work/W2891874693> ?p ?o ?g. }
Showing items 1 to 79 of
79
with 100 items per page.
- W2891874693 endingPage "4612" @default.
- W2891874693 startingPage "4601" @default.
- W2891874693 abstract "Time series classification using deep neural networks, such as convolutional neural networks (CNN), operate on the spectral decomposition of the time series computed using a preprocessing step. This step can include a large number of hyperparameters, such as window length, filter widths, and filter shapes, each with a range of possible values that must be chosen using time and data intensive cross-validation procedures. We propose the wavelet deconvolution (WD) layer as an efficient alternative to this preprocessing step that eliminates a significant number of hyperparameters. The WD layer uses wavelet functions with adjustable scale parameters to learn the spectral decomposition directly from the signal. Using backpropagation, we show the scale parameters can be optimized with gradient descent. Furthermore, the WD layer adds interpretability to the learned time series classifier by exploiting the properties of the wavelet transform. In our experiments, we show that the WD layer can automatically extract the frequency content used to generate a dataset. The WD layer combined with a CNN applied to the phone recognition task on the TIMIT database achieves a phone error rate of 18.1%, a relative improvement of 4% over the baseline CNN. Experiments on a dataset where engineered features are not available showed WD+CNN is the best performing method. Our results show that the WD layer can improve neural network based time series classifiers both in accuracy and interpretability by learning directly from the input signal." @default.
- W2891874693 created "2018-09-27" @default.
- W2891874693 creator A5073676361 @default.
- W2891874693 creator A5083382561 @default.
- W2891874693 date "2018-01-01" @default.
- W2891874693 modified "2023-09-24" @default.
- W2891874693 title "Learning filter widths of spectral decompositions with wavelets" @default.
- W2891874693 hasPublicationYear "2018" @default.
- W2891874693 type Work @default.
- W2891874693 sameAs 2891874693 @default.
- W2891874693 citedByCount "4" @default.
- W2891874693 countsByYear W28918746932019 @default.
- W2891874693 countsByYear W28918746932020 @default.
- W2891874693 countsByYear W28918746932021 @default.
- W2891874693 crossrefType "proceedings-article" @default.
- W2891874693 hasAuthorship W2891874693A5073676361 @default.
- W2891874693 hasAuthorship W2891874693A5083382561 @default.
- W2891874693 hasConcept C106131492 @default.
- W2891874693 hasConcept C108583219 @default.
- W2891874693 hasConcept C11413529 @default.
- W2891874693 hasConcept C153180895 @default.
- W2891874693 hasConcept C154945302 @default.
- W2891874693 hasConcept C155032097 @default.
- W2891874693 hasConcept C174576160 @default.
- W2891874693 hasConcept C22019652 @default.
- W2891874693 hasConcept C23224414 @default.
- W2891874693 hasConcept C2778724510 @default.
- W2891874693 hasConcept C2781067378 @default.
- W2891874693 hasConcept C31972630 @default.
- W2891874693 hasConcept C41008148 @default.
- W2891874693 hasConcept C47432892 @default.
- W2891874693 hasConcept C50644808 @default.
- W2891874693 hasConcept C81363708 @default.
- W2891874693 hasConceptScore W2891874693C106131492 @default.
- W2891874693 hasConceptScore W2891874693C108583219 @default.
- W2891874693 hasConceptScore W2891874693C11413529 @default.
- W2891874693 hasConceptScore W2891874693C153180895 @default.
- W2891874693 hasConceptScore W2891874693C154945302 @default.
- W2891874693 hasConceptScore W2891874693C155032097 @default.
- W2891874693 hasConceptScore W2891874693C174576160 @default.
- W2891874693 hasConceptScore W2891874693C22019652 @default.
- W2891874693 hasConceptScore W2891874693C23224414 @default.
- W2891874693 hasConceptScore W2891874693C2778724510 @default.
- W2891874693 hasConceptScore W2891874693C2781067378 @default.
- W2891874693 hasConceptScore W2891874693C31972630 @default.
- W2891874693 hasConceptScore W2891874693C41008148 @default.
- W2891874693 hasConceptScore W2891874693C47432892 @default.
- W2891874693 hasConceptScore W2891874693C50644808 @default.
- W2891874693 hasConceptScore W2891874693C81363708 @default.
- W2891874693 hasLocation W28918746931 @default.
- W2891874693 hasOpenAccess W2891874693 @default.
- W2891874693 hasPrimaryLocation W28918746931 @default.
- W2891874693 hasRelatedWork W1977426725 @default.
- W2891874693 hasRelatedWork W1997636678 @default.
- W2891874693 hasRelatedWork W2068359377 @default.
- W2891874693 hasRelatedWork W2400676773 @default.
- W2891874693 hasRelatedWork W2765547645 @default.
- W2891874693 hasRelatedWork W2769143033 @default.
- W2891874693 hasRelatedWork W2956033525 @default.
- W2891874693 hasRelatedWork W2963838685 @default.
- W2891874693 hasRelatedWork W2968460295 @default.
- W2891874693 hasRelatedWork W2969333702 @default.
- W2891874693 hasRelatedWork W2980669912 @default.
- W2891874693 hasRelatedWork W3036550712 @default.
- W2891874693 hasRelatedWork W3040402464 @default.
- W2891874693 hasRelatedWork W3048957648 @default.
- W2891874693 hasRelatedWork W3099434629 @default.
- W2891874693 hasRelatedWork W3103740627 @default.
- W2891874693 hasRelatedWork W3117446604 @default.
- W2891874693 hasRelatedWork W3211253523 @default.
- W2891874693 hasRelatedWork W3073332169 @default.
- W2891874693 hasRelatedWork W3086722525 @default.
- W2891874693 hasVolume "31" @default.
- W2891874693 isParatext "false" @default.
- W2891874693 isRetracted "false" @default.
- W2891874693 magId "2891874693" @default.
- W2891874693 workType "article" @default.