FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3121377148> ?p ?o ?g. }

• W3121377148 abstract "Neural networks are sensitive to hyper-parameter and architecture choices. Automated Machine Learning (AutoML) is a promising paradigm for automating these choices. Current ML software libraries, however, are quite limited in handling the dynamic interactions among the components of AutoML. For example, efficientNAS algorithms, such as ENAS and DARTS, typically require an implementation coupling between the search space and search algorithm, the two key components in AutoML. Furthermore, implementing a complex search flow, such as searching architectures within a loop of searching hardware configurations, is difficult. To summarize, changing the search space, search algorithm, or search flow in current ML libraries usually requires a significant change in the program logic. In this paper, we introduce a new way of programming AutoML based on symbolic programming. Under this paradigm, ML programs are mutable, thus can be manipulated easily by another program. As a result, AutoML can be reformulated as an automated process of symbolic manipulation. With this formulation, we decouple the triangle of the search algorithm, the search space and the child program. This decoupling makes it easy to change the search space and search algorithm (without and with weight sharing), as well as to add search capabilities to existing code and implement complex search flows. We then introduce PyGlove, a new Python library that implements this paradigm. Through case studies on ImageNet and NAS-Bench-101, we show that with PyGlove users can easily convert a static program into a search space, quickly iterate on the search spaces and search algorithms, and craft complex search flows to achieve better results." @default.
• W3121377148 created "2021-02-01" @default.
• W3121377148 creator A5012225004 @default.
• W3121377148 creator A5038861110 @default.
• W3121377148 creator A5049565925 @default.
• W3121377148 creator A5049581744 @default.
• W3121377148 creator A5069653714 @default.
• W3121377148 creator A5069705467 @default.
• W3121377148 creator A5069837273 @default.
• W3121377148 creator A5071501764 @default.
• W3121377148 creator A5085154966 @default.
• W3121377148 creator A5088551093 @default.
• W3121377148 date "2021-01-21" @default.
• W3121377148 modified "2023-10-16" @default.
• W3121377148 title "PyGlove: Symbolic Programming for Automated Machine Learning" @default.
• W3121377148 cites W1590649805 @default.
• W3121377148 cites W2010331618 @default.
• W3121377148 cites W2045255985 @default.
• W3121377148 cites W2060610732 @default.
• W3121377148 cites W2089674328 @default.
• W3121377148 cites W2097058024 @default.
• W3121377148 cites W2097998348 @default.
• W3121377148 cites W2105380246 @default.
• W3121377148 cites W2119717200 @default.
• W3121377148 cites W2131241448 @default.
• W3121377148 cites W2155893237 @default.
• W3121377148 cites W2162960800 @default.
• W3121377148 cites W2163605009 @default.
• W3121377148 cites W2173702800 @default.
• W3121377148 cites W2182361439 @default.
• W3121377148 cites W2402144811 @default.
• W3121377148 cites W2419597278 @default.
• W3121377148 cites W2550100435 @default.
• W3121377148 cites W2550120381 @default.
• W3121377148 cites W2594529350 @default.
• W3121377148 cites W2601273560 @default.
• W3121377148 cites W2608595939 @default.
• W3121377148 cites W2732547613 @default.
• W3121377148 cites W2810075754 @default.
• W3121377148 cites W2883188678 @default.
• W3121377148 cites W2891991579 @default.
• W3121377148 cites W2898586683 @default.
• W3121377148 cites W2899971035 @default.
• W3121377148 cites W2905672847 @default.
• W3121377148 cites W2916118939 @default.
• W3121377148 cites W2928941594 @default.
• W3121377148 cites W2936599103 @default.
• W3121377148 cites W2945790622 @default.
• W3121377148 cites W2949676527 @default.
• W3121377148 cites W2955425717 @default.
• W3121377148 cites W2962746461 @default.
• W3121377148 cites W2963163009 @default.
• W3121377148 cites W2963374479 @default.
• W3121377148 cites W2963748792 @default.
• W3121377148 cites W2963918968 @default.
• W3121377148 cites W2963929497 @default.
• W3121377148 cites W2964024268 @default.
• W3121377148 cites W2964081807 @default.
• W3121377148 cites W2964259004 @default.
• W3121377148 cites W2965658867 @default.
• W3121377148 cites W2967733054 @default.
• W3121377148 cites W2970971581 @default.
• W3121377148 cites W2971001037 @default.
• W3121377148 cites W2976890614 @default.
• W3121377148 cites W2980137827 @default.
• W3121377148 cites W3011307848 @default.
• W3121377148 cites W3035400692 @default.
• W3121377148 cites W3082154327 @default.
• W3121377148 cites W3092207651 @default.
• W3121377148 cites W3094801149 @default.
• W3121377148 cites W3117590783 @default.
• W3121377148 doi "https://doi.org/10.48550/arxiv.2101.08809" @default.
• W3121377148 hasPublicationYear "2021" @default.
• W3121377148 type Work @default.
• W3121377148 sameAs 3121377148 @default.
• W3121377148 citedByCount "2" @default.
• W3121377148 countsByYear W31213771482021 @default.
• W3121377148 crossrefType "posted-content" @default.
• W3121377148 hasAuthorship W3121377148A5012225004 @default.
• W3121377148 hasAuthorship W3121377148A5038861110 @default.
• W3121377148 hasAuthorship W3121377148A5049565925 @default.
• W3121377148 hasAuthorship W3121377148A5049581744 @default.
• W3121377148 hasAuthorship W3121377148A5069653714 @default.
• W3121377148 hasAuthorship W3121377148A5069705467 @default.
• W3121377148 hasAuthorship W3121377148A5069837273 @default.
• W3121377148 hasAuthorship W3121377148A5071501764 @default.
• W3121377148 hasAuthorship W3121377148A5085154966 @default.
• W3121377148 hasAuthorship W3121377148A5088551093 @default.
• W3121377148 hasBestOaLocation W31213771481 @default.
• W3121377148 hasConcept C119857082 @default.
• W3121377148 hasConcept C125583679 @default.
• W3121377148 hasConcept C154945302 @default.
• W3121377148 hasConcept C19889080 @default.
• W3121377148 hasConcept C199360897 @default.
• W3121377148 hasConcept C203208320 @default.
• W3121377148 hasConcept C34165917 @default.
• W3121377148 hasConcept C41008148 @default.
• W3121377148 hasConcept C50033165 @default.
• W3121377148 hasConcept C519991488 @default.
• W3121377148 hasConcept C61714564 @default.

• next