FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1673391364> ?p ?o ?g. }

• W1673391364 abstract "In this paper, we propose Multi-state Activation Functions (MSAFs) for Deep Neural Networks (DNNs). These multi-state functions do extra classification based on the 2-state Logistic function. Discussions on the MSAFs reveal that these activation functions have potentials for altering the parameter distribution of the DNN models, improving model performances and reducing model sizes. Meanwhile, an extension of the XOR problem indicates how neural networks with the multistate functions facilitate classifying patterns. Furthermore, basing on running average mean-normalisation rules, we actualise a combination of mean-normalised optimisation with the MSAFs as well as Singular Value Decomposition (SVD). Experimental results on TIMIT reveal that acoustic models based on DNNs can be improved by applying the MSAFs. The models obtain better phone error rates when the Logistic function is replaced with the multi-state functions. Further experiments on large vocabulary continuous speech recognition tasks reveal that the MSAFs and mean-normalised Stochastic Gradient Descent (MN-SGD) bring better recognition performances for DNNs in comparison with the conventional Logistic function and SGD learning method. Beyond this, the combination of the MSAFs, the SVD method and MN-SGD shrinks the parameter scales of DNNs to 44% approximately, leading to considerable increasing on decoding speed and decreasing on model sizes without any loss of recognition performances." @default.
• W1673391364 created "2016-06-24" @default.
• W1673391364 creator A5022188084 @default.
• W1673391364 creator A5026226560 @default.
• W1673391364 creator A5037325282 @default.
• W1673391364 creator A5051330675 @default.
• W1673391364 date "2015-07-01" @default.
• W1673391364 modified "2023-09-25" @default.
• W1673391364 title "A Combination of Multi-state Activation Functions, Mean-normalisation and Singular Value Decomposition for learning Deep Neural Networks" @default.
• W1673391364 cites W128171219 @default.
• W1673391364 cites W1499864241 @default.
• W1673391364 cites W1593564458 @default.
• W1673391364 cites W196761320 @default.
• W1673391364 cites W1980474876 @default.
• W1673391364 cites W1990532266 @default.
• W1673391364 cites W1993882792 @default.
• W1673391364 cites W2014506068 @default.
• W1673391364 cites W2035424729 @default.
• W1673391364 cites W2056738732 @default.
• W1673391364 cites W2064750458 @default.
• W1673391364 cites W2084894614 @default.
• W1673391364 cites W2084910356 @default.
• W1673391364 cites W2087402357 @default.
• W1673391364 cites W2091432990 @default.
• W1673391364 cites W2095168618 @default.
• W1673391364 cites W2136922672 @default.
• W1673391364 cites W2143612262 @default.
• W1673391364 cites W2147768505 @default.
• W1673391364 cites W2168231600 @default.
• W1673391364 cites W2294543795 @default.
• W1673391364 cites W2296748324 @default.
• W1673391364 cites W2403040875 @default.
• W1673391364 cites W71081281 @default.
• W1673391364 cites W1990471071 @default.
• W1673391364 doi "https://doi.org/10.1109/ijcnn.2015.7280321" @default.
• W1673391364 hasPublicationYear "2015" @default.
• W1673391364 type Work @default.
• W1673391364 sameAs 1673391364 @default.
• W1673391364 citedByCount "1" @default.
• W1673391364 countsByYear W16733913642018 @default.
• W1673391364 crossrefType "proceedings-article" @default.
• W1673391364 hasAuthorship W1673391364A5022188084 @default.
• W1673391364 hasAuthorship W1673391364A5026226560 @default.
• W1673391364 hasAuthorship W1673391364A5037325282 @default.
• W1673391364 hasAuthorship W1673391364A5051330675 @default.
• W1673391364 hasConcept C11413529 @default.
• W1673391364 hasConcept C134306372 @default.
• W1673391364 hasConcept C14036430 @default.
• W1673391364 hasConcept C153180895 @default.
• W1673391364 hasConcept C154945302 @default.
• W1673391364 hasConcept C164660894 @default.
• W1673391364 hasConcept C206688291 @default.
• W1673391364 hasConcept C22789450 @default.
• W1673391364 hasConcept C23224414 @default.
• W1673391364 hasConcept C2778724510 @default.
• W1673391364 hasConcept C33923547 @default.
• W1673391364 hasConcept C38365724 @default.
• W1673391364 hasConcept C40969351 @default.
• W1673391364 hasConcept C41008148 @default.
• W1673391364 hasConcept C48103436 @default.
• W1673391364 hasConcept C50644808 @default.
• W1673391364 hasConcept C57273362 @default.
• W1673391364 hasConcept C78458016 @default.
• W1673391364 hasConcept C86803240 @default.
• W1673391364 hasConceptScore W1673391364C11413529 @default.
• W1673391364 hasConceptScore W1673391364C134306372 @default.
• W1673391364 hasConceptScore W1673391364C14036430 @default.
• W1673391364 hasConceptScore W1673391364C153180895 @default.
• W1673391364 hasConceptScore W1673391364C154945302 @default.
• W1673391364 hasConceptScore W1673391364C164660894 @default.
• W1673391364 hasConceptScore W1673391364C206688291 @default.
• W1673391364 hasConceptScore W1673391364C22789450 @default.
• W1673391364 hasConceptScore W1673391364C23224414 @default.
• W1673391364 hasConceptScore W1673391364C2778724510 @default.
• W1673391364 hasConceptScore W1673391364C33923547 @default.
• W1673391364 hasConceptScore W1673391364C38365724 @default.
• W1673391364 hasConceptScore W1673391364C40969351 @default.
• W1673391364 hasConceptScore W1673391364C41008148 @default.
• W1673391364 hasConceptScore W1673391364C48103436 @default.
• W1673391364 hasConceptScore W1673391364C50644808 @default.
• W1673391364 hasConceptScore W1673391364C57273362 @default.
• W1673391364 hasConceptScore W1673391364C78458016 @default.
• W1673391364 hasConceptScore W1673391364C86803240 @default.
• W1673391364 hasLocation W16733913641 @default.
• W1673391364 hasOpenAccess W1673391364 @default.
• W1673391364 hasPrimaryLocation W16733913641 @default.
• W1673391364 hasRelatedWork W1502970866 @default.
• W1673391364 hasRelatedWork W1583347319 @default.
• W1673391364 hasRelatedWork W1590749351 @default.
• W1673391364 hasRelatedWork W1593539490 @default.
• W1673391364 hasRelatedWork W1600107960 @default.
• W1673391364 hasRelatedWork W2007295947 @default.
• W1673391364 hasRelatedWork W2023795513 @default.
• W1673391364 hasRelatedWork W2095191728 @default.
• W1673391364 hasRelatedWork W2117596213 @default.
• W1673391364 hasRelatedWork W2155079222 @default.
• W1673391364 hasRelatedWork W2162962395 @default.
• W1673391364 hasRelatedWork W2166459386 @default.
• W1673391364 hasRelatedWork W2353091813 @default.
• W1673391364 hasRelatedWork W2512885195 @default.

• next