Matches in SemOpenAlex for { <https://semopenalex.org/work/W2246134370> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W2246134370 abstract "Restricted Boltzmann Machines (RBMs) are generative models which can learn useful representations from samples of a dataset in an unsupervised fashion. They have been widely employed as an unsupervised pre-training method in machine learning. RBMs have been modified to model time series in two main ways: The Temporal RBM stacks a number of RBMs laterally and introduces temporal dependencies between the hidden layer units; The Conditional RBM, on the other hand, considers past samples of the dataset as a conditional bias and learns a representation which takes these into account. Here we propose a new training method for both the TRBM and the CRBM, which enforces the dynamic structure of temporal datasets. We do so by treating the temporal models as denoising autoencoders, considering past frames of the dataset as corrupted versions of the present frame and minimizing the reconstruction error of the present data by the model. We call this approach Temporal Autoencoding. This leads to a significant improvement in the performance of both models in a filling-in-frames task across a number of datasets. The error reduction for motion capture data is 56% for the CRBM and 80% for the TRBM. Taking the posterior mean prediction instead of single samples further improves the model's estimates, decreasing the error by as much as 91% for the CRBM on motion capture data. We also trained the model to perform forecasting on a large number of datasets and have found TA pretraining to consistently improve the performance of the forecasts. Furthermore, by looking at the prediction error across time, we can see that this improvement reflects a better representation of the dynamics of the data as opposed to a bias towards reconstructing the observed data on a short time scale." @default.
- W2246134370 created "2016-06-24" @default.
- W2246134370 creator A5006626558 @default.
- W2246134370 creator A5042543744 @default.
- W2246134370 creator A5071014290 @default.
- W2246134370 creator A5078375150 @default.
- W2246134370 date "2013-09-12" @default.
- W2246134370 modified "2023-10-01" @default.
- W2246134370 title "Temporal Autoencoding Improves Generative Models of Time Series." @default.
- W2246134370 cites W2048665112 @default.
- W2246134370 cites W2100495367 @default.
- W2246134370 cites W2134842679 @default.
- W2246134370 cites W2135341757 @default.
- W2246134370 cites W2145094598 @default.
- W2246134370 cites W2158164339 @default.
- W2246134370 cites W2464714129 @default.
- W2246134370 cites W66838807 @default.
- W2246134370 hasPublicationYear "2013" @default.
- W2246134370 type Work @default.
- W2246134370 sameAs 2246134370 @default.
- W2246134370 citedByCount "3" @default.
- W2246134370 countsByYear W22461343702015 @default.
- W2246134370 countsByYear W22461343702016 @default.
- W2246134370 crossrefType "posted-content" @default.
- W2246134370 hasAuthorship W2246134370A5006626558 @default.
- W2246134370 hasAuthorship W2246134370A5042543744 @default.
- W2246134370 hasAuthorship W2246134370A5071014290 @default.
- W2246134370 hasAuthorship W2246134370A5078375150 @default.
- W2246134370 hasConcept C104114177 @default.
- W2246134370 hasConcept C108583219 @default.
- W2246134370 hasConcept C119857082 @default.
- W2246134370 hasConcept C126042441 @default.
- W2246134370 hasConcept C143724316 @default.
- W2246134370 hasConcept C151730666 @default.
- W2246134370 hasConcept C153180895 @default.
- W2246134370 hasConcept C154945302 @default.
- W2246134370 hasConcept C167966045 @default.
- W2246134370 hasConcept C17744445 @default.
- W2246134370 hasConcept C192576344 @default.
- W2246134370 hasConcept C199354608 @default.
- W2246134370 hasConcept C199539241 @default.
- W2246134370 hasConcept C2776359362 @default.
- W2246134370 hasConcept C39890363 @default.
- W2246134370 hasConcept C41008148 @default.
- W2246134370 hasConcept C59404180 @default.
- W2246134370 hasConcept C76155785 @default.
- W2246134370 hasConcept C8038995 @default.
- W2246134370 hasConcept C86803240 @default.
- W2246134370 hasConcept C94625758 @default.
- W2246134370 hasConceptScore W2246134370C104114177 @default.
- W2246134370 hasConceptScore W2246134370C108583219 @default.
- W2246134370 hasConceptScore W2246134370C119857082 @default.
- W2246134370 hasConceptScore W2246134370C126042441 @default.
- W2246134370 hasConceptScore W2246134370C143724316 @default.
- W2246134370 hasConceptScore W2246134370C151730666 @default.
- W2246134370 hasConceptScore W2246134370C153180895 @default.
- W2246134370 hasConceptScore W2246134370C154945302 @default.
- W2246134370 hasConceptScore W2246134370C167966045 @default.
- W2246134370 hasConceptScore W2246134370C17744445 @default.
- W2246134370 hasConceptScore W2246134370C192576344 @default.
- W2246134370 hasConceptScore W2246134370C199354608 @default.
- W2246134370 hasConceptScore W2246134370C199539241 @default.
- W2246134370 hasConceptScore W2246134370C2776359362 @default.
- W2246134370 hasConceptScore W2246134370C39890363 @default.
- W2246134370 hasConceptScore W2246134370C41008148 @default.
- W2246134370 hasConceptScore W2246134370C59404180 @default.
- W2246134370 hasConceptScore W2246134370C76155785 @default.
- W2246134370 hasConceptScore W2246134370C8038995 @default.
- W2246134370 hasConceptScore W2246134370C86803240 @default.
- W2246134370 hasConceptScore W2246134370C94625758 @default.
- W2246134370 hasLocation W22461343701 @default.
- W2246134370 hasOpenAccess W2246134370 @default.
- W2246134370 hasPrimaryLocation W22461343701 @default.
- W2246134370 hasRelatedWork W2185528074 @default.
- W2246134370 hasRelatedWork W2187294194 @default.
- W2246134370 hasRelatedWork W2394569251 @default.
- W2246134370 hasRelatedWork W2395018931 @default.
- W2246134370 hasRelatedWork W2562973942 @default.
- W2246134370 hasRelatedWork W2605216337 @default.
- W2246134370 hasRelatedWork W2763994000 @default.
- W2246134370 hasRelatedWork W2766082768 @default.
- W2246134370 hasRelatedWork W2773850097 @default.
- W2246134370 hasRelatedWork W2783883100 @default.
- W2246134370 hasRelatedWork W2810784658 @default.
- W2246134370 hasRelatedWork W2898843852 @default.
- W2246134370 hasRelatedWork W2948012784 @default.
- W2246134370 hasRelatedWork W2966819461 @default.
- W2246134370 hasRelatedWork W2981654920 @default.
- W2246134370 hasRelatedWork W3008659735 @default.
- W2246134370 hasRelatedWork W3019210193 @default.
- W2246134370 hasRelatedWork W3034207266 @default.
- W2246134370 hasRelatedWork W3036996956 @default.
- W2246134370 hasRelatedWork W3049554879 @default.
- W2246134370 isParatext "false" @default.
- W2246134370 isRetracted "false" @default.
- W2246134370 magId "2246134370" @default.
- W2246134370 workType "article" @default.