Matches in SemOpenAlex for { <https://semopenalex.org/work/W2602428493> ?p ?o ?g. }
- W2602428493 endingPage "52" @default.
- W2602428493 startingPage "41" @default.
- W2602428493 abstract "To aggregate diverse learners and to train deep architectures are the two principal avenues towards increasing the expressive capabilities of neural networks. Therefore, their combinations merit attention. In this contribution, we study how to apply some conventional diversity methods –bagging and label switching– to a general deep machine, the stacked denoising auto-encoding classifier, in order to solve a number of appropriately selected image recognition problems. The main conclusion of our work is that binarizing multi-class problems is the key to obtain benefit from those diversity methods. Additionally, we check that adding other kinds of performance improvement procedures, such as pre-emphasizing training samples and elastic distortion mechanisms, further increases the quality of the results. In particular, an appropriate combination of all the above methods leads us to reach a new absolute record in classifying MNIST handwritten digits. These facts reveal that there are clear opportunities for designing more powerful classifiers by means of combining different improvement techniques." @default.
- W2602428493 created "2017-04-07" @default.
- W2602428493 creator A5046532092 @default.
- W2602428493 creator A5084423849 @default.
- W2602428493 date "2018-01-01" @default.
- W2602428493 modified "2023-09-26" @default.
- W2602428493 title "On building ensembles of stacked denoising auto-encoding classifiers and their further improvement" @default.
- W2602428493 cites W1580948147 @default.
- W2602428493 cites W1623615678 @default.
- W2602428493 cites W1676820704 @default.
- W2602428493 cites W1978508564 @default.
- W2602428493 cites W1988790447 @default.
- W2602428493 cites W1994410331 @default.
- W2602428493 cites W2019575783 @default.
- W2602428493 cites W2033272835 @default.
- W2602428493 cites W2033337358 @default.
- W2602428493 cites W2038705219 @default.
- W2602428493 cites W2061119986 @default.
- W2602428493 cites W2061913380 @default.
- W2602428493 cites W2068238590 @default.
- W2602428493 cites W2072181174 @default.
- W2602428493 cites W2079935202 @default.
- W2602428493 cites W2090293612 @default.
- W2602428493 cites W2092530862 @default.
- W2602428493 cites W2093559122 @default.
- W2602428493 cites W2095955281 @default.
- W2602428493 cites W2103496339 @default.
- W2602428493 cites W2107998050 @default.
- W2602428493 cites W2112796928 @default.
- W2602428493 cites W2113853244 @default.
- W2602428493 cites W2116374865 @default.
- W2602428493 cites W2118023920 @default.
- W2602428493 cites W2125679713 @default.
- W2602428493 cites W2132407367 @default.
- W2602428493 cites W2133721713 @default.
- W2602428493 cites W2136922672 @default.
- W2602428493 cites W2137983211 @default.
- W2602428493 cites W2147800946 @default.
- W2602428493 cites W2150884987 @default.
- W2602428493 cites W2163922914 @default.
- W2602428493 cites W2166782149 @default.
- W2602428493 cites W2342792901 @default.
- W2602428493 cites W2564734887 @default.
- W2602428493 cites W2911964244 @default.
- W2602428493 cites W4205947740 @default.
- W2602428493 cites W4212883601 @default.
- W2602428493 cites W4231109964 @default.
- W2602428493 cites W4244952642 @default.
- W2602428493 doi "https://doi.org/10.1016/j.inffus.2017.03.008" @default.
- W2602428493 hasPublicationYear "2018" @default.
- W2602428493 type Work @default.
- W2602428493 sameAs 2602428493 @default.
- W2602428493 citedByCount "31" @default.
- W2602428493 countsByYear W26024284932017 @default.
- W2602428493 countsByYear W26024284932018 @default.
- W2602428493 countsByYear W26024284932019 @default.
- W2602428493 countsByYear W26024284932020 @default.
- W2602428493 countsByYear W26024284932021 @default.
- W2602428493 countsByYear W26024284932022 @default.
- W2602428493 countsByYear W26024284932023 @default.
- W2602428493 crossrefType "journal-article" @default.
- W2602428493 hasAuthorship W2602428493A5046532092 @default.
- W2602428493 hasAuthorship W2602428493A5084423849 @default.
- W2602428493 hasConcept C119857082 @default.
- W2602428493 hasConcept C125411270 @default.
- W2602428493 hasConcept C153180895 @default.
- W2602428493 hasConcept C154945302 @default.
- W2602428493 hasConcept C163294075 @default.
- W2602428493 hasConcept C190502265 @default.
- W2602428493 hasConcept C41008148 @default.
- W2602428493 hasConcept C50644808 @default.
- W2602428493 hasConcept C95623464 @default.
- W2602428493 hasConceptScore W2602428493C119857082 @default.
- W2602428493 hasConceptScore W2602428493C125411270 @default.
- W2602428493 hasConceptScore W2602428493C153180895 @default.
- W2602428493 hasConceptScore W2602428493C154945302 @default.
- W2602428493 hasConceptScore W2602428493C163294075 @default.
- W2602428493 hasConceptScore W2602428493C190502265 @default.
- W2602428493 hasConceptScore W2602428493C41008148 @default.
- W2602428493 hasConceptScore W2602428493C50644808 @default.
- W2602428493 hasConceptScore W2602428493C95623464 @default.
- W2602428493 hasFunder F4320321837 @default.
- W2602428493 hasFunder F4320335166 @default.
- W2602428493 hasLocation W26024284931 @default.
- W2602428493 hasOpenAccess W2602428493 @default.
- W2602428493 hasPrimaryLocation W26024284931 @default.
- W2602428493 hasRelatedWork W2001652754 @default.
- W2602428493 hasRelatedWork W2549006548 @default.
- W2602428493 hasRelatedWork W2810865670 @default.
- W2602428493 hasRelatedWork W2961085424 @default.
- W2602428493 hasRelatedWork W3156786002 @default.
- W2602428493 hasRelatedWork W4214932115 @default.
- W2602428493 hasRelatedWork W4221015625 @default.
- W2602428493 hasRelatedWork W4243326899 @default.
- W2602428493 hasRelatedWork W1629725936 @default.
- W2602428493 hasRelatedWork W3158004940 @default.
- W2602428493 hasVolume "39" @default.
- W2602428493 isParatext "false" @default.