Matches in SemOpenAlex for { <https://semopenalex.org/work/W113669206> ?p ?o ?g. }
- W113669206 abstract "Unsupervised feature learning algorithms based on convolutional formulations of independent components analysis (ICA) have been demonstrated to yield state-of-the-art results in several action recognition benchmarks. However, existing approaches do not allow for the number of latent components (features) to be automatically inferred from the data in an unsupervised manner. This is a significant disadvantage of the state-of-the-art, as it results in considerable burden imposed on researchers and practitioners, who must resort to tedious cross-validation procedures to obtain the optimal number of latent features. To resolve these issues, in this paper we introduce a convolutional nonparametric Bayesian sparse ICA architecture for overcomplete feature learning from high-dimensional data. Our method utilizes an Indian buffet process prior to facilitate inference of the appropriate number of latent features under a hybrid variational inference algorithm, scalable to massive datasets. As we show, our model can be naturally used to obtain deep unsupervised hierarchical feature extractors, by greedily stacking successive model layers, similar to existing approaches. In addition, inference for this model is completely heuristics-free, thus, it obviates the need of tedious parameter tuning, which is a major challenge most deep learning approaches are faced with. We evaluate our method on several action recognition benchmarks, and exhibit its advantages over the state-of-the-art." @default.
- W113669206 created "2016-06-24" @default.
- W113669206 creator A5021338623 @default.
- W113669206 creator A5040795718 @default.
- W113669206 date "2015-12-01" @default.
- W113669206 modified "2023-09-26" @default.
- W113669206 title "A Nonparametric Bayesian Approach toward Stacked Convolutional Independent Component Analysis" @default.
- W113669206 cites W104847522 @default.
- W113669206 cites W1516111018 @default.
- W113669206 cites W1534763723 @default.
- W113669206 cites W1586730761 @default.
- W113669206 cites W1625086225 @default.
- W113669206 cites W1663973292 @default.
- W113669206 cites W1963995223 @default.
- W113669206 cites W1993229407 @default.
- W113669206 cites W1999192586 @default.
- W113669206 cites W1999974018 @default.
- W113669206 cites W2016053056 @default.
- W113669206 cites W2024868105 @default.
- W113669206 cites W2038885294 @default.
- W113669206 cites W2068730032 @default.
- W113669206 cites W2091662199 @default.
- W113669206 cites W2096691069 @default.
- W113669206 cites W2100916003 @default.
- W113669206 cites W2106813372 @default.
- W113669206 cites W2112274848 @default.
- W113669206 cites W2113606819 @default.
- W113669206 cites W2120432001 @default.
- W113669206 cites W2126574503 @default.
- W113669206 cites W2126658427 @default.
- W113669206 cites W2128002512 @default.
- W113669206 cites W2130325614 @default.
- W113669206 cites W2136922672 @default.
- W113669206 cites W2145094598 @default.
- W113669206 cites W2148603752 @default.
- W113669206 cites W2157514233 @default.
- W113669206 cites W2161381512 @default.
- W113669206 cites W2163292664 @default.
- W113669206 cites W2171278750 @default.
- W113669206 cites W2217045154 @default.
- W113669206 cites W237546731 @default.
- W113669206 cites W2533739470 @default.
- W113669206 doi "https://doi.org/10.1109/iccv.2015.321" @default.
- W113669206 hasPublicationYear "2015" @default.
- W113669206 type Work @default.
- W113669206 sameAs 113669206 @default.
- W113669206 citedByCount "4" @default.
- W113669206 countsByYear W1136692062015 @default.
- W113669206 countsByYear W1136692062017 @default.
- W113669206 countsByYear W1136692062021 @default.
- W113669206 crossrefType "proceedings-article" @default.
- W113669206 hasAuthorship W113669206A5021338623 @default.
- W113669206 hasAuthorship W113669206A5040795718 @default.
- W113669206 hasBestOaLocation W1136692062 @default.
- W113669206 hasConcept C107673813 @default.
- W113669206 hasConcept C119857082 @default.
- W113669206 hasConcept C138885662 @default.
- W113669206 hasConcept C153180895 @default.
- W113669206 hasConcept C154945302 @default.
- W113669206 hasConcept C160234255 @default.
- W113669206 hasConcept C2776214188 @default.
- W113669206 hasConcept C2776401178 @default.
- W113669206 hasConcept C41008148 @default.
- W113669206 hasConcept C41895202 @default.
- W113669206 hasConcept C59404180 @default.
- W113669206 hasConcept C8038995 @default.
- W113669206 hasConcept C81363708 @default.
- W113669206 hasConceptScore W113669206C107673813 @default.
- W113669206 hasConceptScore W113669206C119857082 @default.
- W113669206 hasConceptScore W113669206C138885662 @default.
- W113669206 hasConceptScore W113669206C153180895 @default.
- W113669206 hasConceptScore W113669206C154945302 @default.
- W113669206 hasConceptScore W113669206C160234255 @default.
- W113669206 hasConceptScore W113669206C2776214188 @default.
- W113669206 hasConceptScore W113669206C2776401178 @default.
- W113669206 hasConceptScore W113669206C41008148 @default.
- W113669206 hasConceptScore W113669206C41895202 @default.
- W113669206 hasConceptScore W113669206C59404180 @default.
- W113669206 hasConceptScore W113669206C8038995 @default.
- W113669206 hasConceptScore W113669206C81363708 @default.
- W113669206 hasLocation W1136692061 @default.
- W113669206 hasLocation W1136692062 @default.
- W113669206 hasLocation W1136692063 @default.
- W113669206 hasOpenAccess W113669206 @default.
- W113669206 hasPrimaryLocation W1136692061 @default.
- W113669206 hasRelatedWork W1887168758 @default.
- W113669206 hasRelatedWork W1979999753 @default.
- W113669206 hasRelatedWork W2015648984 @default.
- W113669206 hasRelatedWork W2065143024 @default.
- W113669206 hasRelatedWork W2091964862 @default.
- W113669206 hasRelatedWork W2172081788 @default.
- W113669206 hasRelatedWork W2299646912 @default.
- W113669206 hasRelatedWork W2441742266 @default.
- W113669206 hasRelatedWork W2550521683 @default.
- W113669206 hasRelatedWork W2571196506 @default.
- W113669206 hasRelatedWork W2801572136 @default.
- W113669206 hasRelatedWork W2898826688 @default.
- W113669206 hasRelatedWork W2945232940 @default.
- W113669206 hasRelatedWork W2962984958 @default.
- W113669206 hasRelatedWork W2972135110 @default.