Matches in SemOpenAlex for { <https://semopenalex.org/work/W2891448299> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W2891448299 abstract "Author(s): Xie, Saining | Advisor(s): Tu, Zhuowen | Abstract: With the support of big-data and big-compute, deep learning has reshaped the landscape of research and applications in artificial intelligence. Whilst traditional hand-guided feature engineering in many cases is simplified, the deep network architectures become increasingly more complex. A central question is whether we can distill the minimal set of structural priors that can provide us the maximal flexibility, and lead us to richer sets of structural primitives. Those structural priors will make the learning process more effective, and potentially lay the foundations towards the ultimate goal of building general intelligent systems. This dissertation focuses on how we can tackle different real world problems in computer vision and machine learning with carefully designed neural network architectures, guided by simple yet effective structural priors. In particular, this thesis focuses on two structural priors that have proven to be useful and generalizable in many different scenarios: the multi-scale prior, with an application in edge detection, and the sparse-connectivity prior implemented for generic visual recognition. Examples will be presented in the last part, on how to learn meaningful structures directly from data, rather than hard-wiring them by, for example, learning a convolutional pseudo-prior in the label space, or adopting a dynamic self-attention mechanism." @default.
- W2891448299 created "2018-09-27" @default.
- W2891448299 creator A5024785349 @default.
- W2891448299 date "2018-01-01" @default.
- W2891448299 modified "2023-09-27" @default.
- W2891448299 title "Deep Representation Learning with Induced Structural Priors" @default.
- W2891448299 hasPublicationYear "2018" @default.
- W2891448299 type Work @default.
- W2891448299 sameAs 2891448299 @default.
- W2891448299 citedByCount "0" @default.
- W2891448299 crossrefType "journal-article" @default.
- W2891448299 hasAuthorship W2891448299A5024785349 @default.
- W2891448299 hasConcept C105795698 @default.
- W2891448299 hasConcept C107673813 @default.
- W2891448299 hasConcept C108583219 @default.
- W2891448299 hasConcept C119857082 @default.
- W2891448299 hasConcept C138885662 @default.
- W2891448299 hasConcept C154945302 @default.
- W2891448299 hasConcept C177264268 @default.
- W2891448299 hasConcept C17744445 @default.
- W2891448299 hasConcept C177769412 @default.
- W2891448299 hasConcept C199360897 @default.
- W2891448299 hasConcept C199539241 @default.
- W2891448299 hasConcept C2776359362 @default.
- W2891448299 hasConcept C2776401178 @default.
- W2891448299 hasConcept C2780598303 @default.
- W2891448299 hasConcept C33923547 @default.
- W2891448299 hasConcept C41008148 @default.
- W2891448299 hasConcept C41895202 @default.
- W2891448299 hasConcept C81363708 @default.
- W2891448299 hasConcept C94625758 @default.
- W2891448299 hasConceptScore W2891448299C105795698 @default.
- W2891448299 hasConceptScore W2891448299C107673813 @default.
- W2891448299 hasConceptScore W2891448299C108583219 @default.
- W2891448299 hasConceptScore W2891448299C119857082 @default.
- W2891448299 hasConceptScore W2891448299C138885662 @default.
- W2891448299 hasConceptScore W2891448299C154945302 @default.
- W2891448299 hasConceptScore W2891448299C177264268 @default.
- W2891448299 hasConceptScore W2891448299C17744445 @default.
- W2891448299 hasConceptScore W2891448299C177769412 @default.
- W2891448299 hasConceptScore W2891448299C199360897 @default.
- W2891448299 hasConceptScore W2891448299C199539241 @default.
- W2891448299 hasConceptScore W2891448299C2776359362 @default.
- W2891448299 hasConceptScore W2891448299C2776401178 @default.
- W2891448299 hasConceptScore W2891448299C2780598303 @default.
- W2891448299 hasConceptScore W2891448299C33923547 @default.
- W2891448299 hasConceptScore W2891448299C41008148 @default.
- W2891448299 hasConceptScore W2891448299C41895202 @default.
- W2891448299 hasConceptScore W2891448299C81363708 @default.
- W2891448299 hasConceptScore W2891448299C94625758 @default.
- W2891448299 hasLocation W28914482991 @default.
- W2891448299 hasOpenAccess W2891448299 @default.
- W2891448299 hasPrimaryLocation W28914482991 @default.
- W2891448299 hasRelatedWork W108981635 @default.
- W2891448299 hasRelatedWork W13303319 @default.
- W2891448299 hasRelatedWork W2107218456 @default.
- W2891448299 hasRelatedWork W2122308921 @default.
- W2891448299 hasRelatedWork W2370160218 @default.
- W2891448299 hasRelatedWork W2527946065 @default.
- W2891448299 hasRelatedWork W2571500919 @default.
- W2891448299 hasRelatedWork W2618724401 @default.
- W2891448299 hasRelatedWork W2705882626 @default.
- W2891448299 hasRelatedWork W2787334611 @default.
- W2891448299 hasRelatedWork W2895835746 @default.
- W2891448299 hasRelatedWork W2906883261 @default.
- W2891448299 hasRelatedWork W2998715769 @default.
- W2891448299 hasRelatedWork W3004632877 @default.
- W2891448299 hasRelatedWork W3013068546 @default.
- W2891448299 hasRelatedWork W3021965466 @default.
- W2891448299 hasRelatedWork W3049451275 @default.
- W2891448299 hasRelatedWork W3136902004 @default.
- W2891448299 hasRelatedWork W3176216676 @default.
- W2891448299 hasRelatedWork W2745032443 @default.
- W2891448299 isParatext "false" @default.
- W2891448299 isRetracted "false" @default.
- W2891448299 magId "2891448299" @default.
- W2891448299 workType "article" @default.