Matches in SemOpenAlex for { <https://semopenalex.org/work/W1833744030> ?p ?o ?g. }
- W1833744030 abstract "In science and engineering, intelligent processing of complex signals such as images, sound or language is often performed by a parameterized hierarchy of nonlinear processing layers, sometimes biologically inspired. Hierarchical systems (or, more generally, nested systems) offer a way to generate complex mappings using simple stages. Each layer performs a different operation and achieves an ever more sophisticated representation of the input, as, for example, in an deep artificial neural network, an object recognition cascade in computer vision or a speech front-end processing. Joint estimation of the parameters of all the layers and selection of an optimal architecture is widely considered to be a difficult numerical nonconvex optimization problem, difficult to parallelize for execution in a distributed computation environment, and requiring significant human expert effort, which leads to suboptimal systems in practice. We describe a general mathematical strategy to learn the parameters and, to some extent, the architecture of nested systems, called the method of auxiliary coordinates (MAC). This replaces the original problem involving a deeply nested function with a constrained problem involving a different function in an augmented space without nesting. The constrained problem may be solved with penalty-based methods using alternating optimization over the parameters and the auxiliary coordinates. MAC has provable convergence, is easy to implement reusing existing algorithms for single layers, can be parallelized trivially and massively, applies even when parameter derivatives are not available or not desirable, and is competitive with state-of-the-art nonlinear optimizers even in the serial computation setting, often providing reasonable models within a few iterations." @default.
- W1833744030 created "2016-06-24" @default.
- W1833744030 creator A5005239198 @default.
- W1833744030 creator A5057957982 @default.
- W1833744030 date "2012-12-24" @default.
- W1833744030 modified "2023-09-27" @default.
- W1833744030 title "Distributed optimization of deeply nested systems" @default.
- W1833744030 cites W1554944419 @default.
- W1833744030 cites W1576278180 @default.
- W1833744030 cites W1578316706 @default.
- W1833744030 cites W1946620893 @default.
- W1833744030 cites W2001141328 @default.
- W1833744030 cites W2015631279 @default.
- W1833744030 cites W2027653851 @default.
- W1833744030 cites W2038497950 @default.
- W1833744030 cites W2049633694 @default.
- W1833744030 cites W2053186076 @default.
- W1833744030 cites W2060895888 @default.
- W1833744030 cites W2099229603 @default.
- W1833744030 cites W2100495367 @default.
- W1833744030 cites W2100659887 @default.
- W1833744030 cites W2105464873 @default.
- W1833744030 cites W2110798204 @default.
- W1833744030 cites W2112796928 @default.
- W1833744030 cites W2125395894 @default.
- W1833744030 cites W2135719924 @default.
- W1833744030 cites W2136111243 @default.
- W1833744030 cites W2139427956 @default.
- W1833744030 cites W2144982973 @default.
- W1833744030 cites W2145889472 @default.
- W1833744030 cites W2149194912 @default.
- W1833744030 cites W2154642048 @default.
- W1833744030 cites W2163490846 @default.
- W1833744030 cites W2164278908 @default.
- W1833744030 cites W2167607759 @default.
- W1833744030 cites W2171490498 @default.
- W1833744030 cites W2185528074 @default.
- W1833744030 cites W2476694670 @default.
- W1833744030 cites W2558164073 @default.
- W1833744030 cites W2963261555 @default.
- W1833744030 cites W3029645440 @default.
- W1833744030 cites W3121926921 @default.
- W1833744030 hasPublicationYear "2012" @default.
- W1833744030 type Work @default.
- W1833744030 sameAs 1833744030 @default.
- W1833744030 citedByCount "11" @default.
- W1833744030 countsByYear W18337440302015 @default.
- W1833744030 countsByYear W18337440302016 @default.
- W1833744030 countsByYear W18337440302017 @default.
- W1833744030 countsByYear W18337440302018 @default.
- W1833744030 countsByYear W18337440302020 @default.
- W1833744030 crossrefType "posted-content" @default.
- W1833744030 hasAuthorship W1833744030A5005239198 @default.
- W1833744030 hasAuthorship W1833744030A5057957982 @default.
- W1833744030 hasConcept C11413529 @default.
- W1833744030 hasConcept C126255220 @default.
- W1833744030 hasConcept C137836250 @default.
- W1833744030 hasConcept C14036430 @default.
- W1833744030 hasConcept C154945302 @default.
- W1833744030 hasConcept C162324750 @default.
- W1833744030 hasConcept C173608175 @default.
- W1833744030 hasConcept C17744445 @default.
- W1833744030 hasConcept C190475519 @default.
- W1833744030 hasConcept C191897082 @default.
- W1833744030 hasConcept C192562407 @default.
- W1833744030 hasConcept C199539241 @default.
- W1833744030 hasConcept C2776359362 @default.
- W1833744030 hasConcept C2776937656 @default.
- W1833744030 hasConcept C31170391 @default.
- W1833744030 hasConcept C33923547 @default.
- W1833744030 hasConcept C34447519 @default.
- W1833744030 hasConcept C41008148 @default.
- W1833744030 hasConcept C45374587 @default.
- W1833744030 hasConcept C50644808 @default.
- W1833744030 hasConcept C78458016 @default.
- W1833744030 hasConcept C80444323 @default.
- W1833744030 hasConcept C86803240 @default.
- W1833744030 hasConcept C94625758 @default.
- W1833744030 hasConceptScore W1833744030C11413529 @default.
- W1833744030 hasConceptScore W1833744030C126255220 @default.
- W1833744030 hasConceptScore W1833744030C137836250 @default.
- W1833744030 hasConceptScore W1833744030C14036430 @default.
- W1833744030 hasConceptScore W1833744030C154945302 @default.
- W1833744030 hasConceptScore W1833744030C162324750 @default.
- W1833744030 hasConceptScore W1833744030C173608175 @default.
- W1833744030 hasConceptScore W1833744030C17744445 @default.
- W1833744030 hasConceptScore W1833744030C190475519 @default.
- W1833744030 hasConceptScore W1833744030C191897082 @default.
- W1833744030 hasConceptScore W1833744030C192562407 @default.
- W1833744030 hasConceptScore W1833744030C199539241 @default.
- W1833744030 hasConceptScore W1833744030C2776359362 @default.
- W1833744030 hasConceptScore W1833744030C2776937656 @default.
- W1833744030 hasConceptScore W1833744030C31170391 @default.
- W1833744030 hasConceptScore W1833744030C33923547 @default.
- W1833744030 hasConceptScore W1833744030C34447519 @default.
- W1833744030 hasConceptScore W1833744030C41008148 @default.
- W1833744030 hasConceptScore W1833744030C45374587 @default.
- W1833744030 hasConceptScore W1833744030C50644808 @default.
- W1833744030 hasConceptScore W1833744030C78458016 @default.
- W1833744030 hasConceptScore W1833744030C80444323 @default.