Matches in SemOpenAlex for { <https://semopenalex.org/work/W4294576319> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W4294576319 abstract "We investigate the complexity of deep neural networks (DNN) that represent piecewise linear (PWL) functions. In particular, we study the number of linear regions, i.e. pieces, that a PWL function represented by a DNN can attain, both theoretically and empirically. We present (i) tighter upper and lower bounds for the maximum number of linear regions on rectifier networks, which are exact for inputs of dimension one; (ii) a first upper bound for multi-layer maxout networks; and (iii) a first method to perform exact enumeration or counting of the number of regions by modeling the DNN with a mixed-integer linear formulation. These bounds come from leveraging the dimension of the space defining each linear region. The results also indicate that a deep rectifier network can only have more linear regions than every shallow counterpart with same number of neurons if that number exceeds the dimension of the input." @default.
- W4294576319 created "2022-09-05" @default.
- W4294576319 creator A5017504613 @default.
- W4294576319 creator A5021227744 @default.
- W4294576319 creator A5060350100 @default.
- W4294576319 date "2017-11-06" @default.
- W4294576319 modified "2023-09-25" @default.
- W4294576319 title "Bounding and Counting Linear Regions of Deep Neural Networks" @default.
- W4294576319 doi "https://doi.org/10.48550/arxiv.1711.02114" @default.
- W4294576319 hasPublicationYear "2017" @default.
- W4294576319 type Work @default.
- W4294576319 citedByCount "0" @default.
- W4294576319 crossrefType "posted-content" @default.
- W4294576319 hasAuthorship W4294576319A5017504613 @default.
- W4294576319 hasAuthorship W4294576319A5021227744 @default.
- W4294576319 hasAuthorship W4294576319A5060350100 @default.
- W4294576319 hasBestOaLocation W42945763191 @default.
- W4294576319 hasConcept C11413529 @default.
- W4294576319 hasConcept C114614502 @default.
- W4294576319 hasConcept C134306372 @default.
- W4294576319 hasConcept C14036430 @default.
- W4294576319 hasConcept C154945302 @default.
- W4294576319 hasConcept C156340839 @default.
- W4294576319 hasConcept C17095337 @default.
- W4294576319 hasConcept C175202392 @default.
- W4294576319 hasConcept C176370821 @default.
- W4294576319 hasConcept C199360897 @default.
- W4294576319 hasConcept C33676613 @default.
- W4294576319 hasConcept C33923547 @default.
- W4294576319 hasConcept C41008148 @default.
- W4294576319 hasConcept C41045048 @default.
- W4294576319 hasConcept C50100734 @default.
- W4294576319 hasConcept C50644808 @default.
- W4294576319 hasConcept C63584917 @default.
- W4294576319 hasConcept C77553402 @default.
- W4294576319 hasConcept C78458016 @default.
- W4294576319 hasConcept C86582703 @default.
- W4294576319 hasConcept C86803240 @default.
- W4294576319 hasConcept C97137487 @default.
- W4294576319 hasConceptScore W4294576319C11413529 @default.
- W4294576319 hasConceptScore W4294576319C114614502 @default.
- W4294576319 hasConceptScore W4294576319C134306372 @default.
- W4294576319 hasConceptScore W4294576319C14036430 @default.
- W4294576319 hasConceptScore W4294576319C154945302 @default.
- W4294576319 hasConceptScore W4294576319C156340839 @default.
- W4294576319 hasConceptScore W4294576319C17095337 @default.
- W4294576319 hasConceptScore W4294576319C175202392 @default.
- W4294576319 hasConceptScore W4294576319C176370821 @default.
- W4294576319 hasConceptScore W4294576319C199360897 @default.
- W4294576319 hasConceptScore W4294576319C33676613 @default.
- W4294576319 hasConceptScore W4294576319C33923547 @default.
- W4294576319 hasConceptScore W4294576319C41008148 @default.
- W4294576319 hasConceptScore W4294576319C41045048 @default.
- W4294576319 hasConceptScore W4294576319C50100734 @default.
- W4294576319 hasConceptScore W4294576319C50644808 @default.
- W4294576319 hasConceptScore W4294576319C63584917 @default.
- W4294576319 hasConceptScore W4294576319C77553402 @default.
- W4294576319 hasConceptScore W4294576319C78458016 @default.
- W4294576319 hasConceptScore W4294576319C86582703 @default.
- W4294576319 hasConceptScore W4294576319C86803240 @default.
- W4294576319 hasConceptScore W4294576319C97137487 @default.
- W4294576319 hasLocation W42945763191 @default.
- W4294576319 hasOpenAccess W4294576319 @default.
- W4294576319 hasPrimaryLocation W42945763191 @default.
- W4294576319 hasRelatedWork W1967853602 @default.
- W4294576319 hasRelatedWork W2001854852 @default.
- W4294576319 hasRelatedWork W2069783012 @default.
- W4294576319 hasRelatedWork W2137609352 @default.
- W4294576319 hasRelatedWork W2616924990 @default.
- W4294576319 hasRelatedWork W2767609493 @default.
- W4294576319 hasRelatedWork W2963241012 @default.
- W4294576319 hasRelatedWork W2963950090 @default.
- W4294576319 hasRelatedWork W3042991734 @default.
- W4294576319 hasRelatedWork W4294576319 @default.
- W4294576319 isParatext "false" @default.
- W4294576319 isRetracted "false" @default.
- W4294576319 workType "article" @default.