FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4383460730> ?p ?o ?g. }

• W4383460730 endingPage "1159" @default.
• W4383460730 startingPage "1073" @default.
• W4383460730 abstract "Machine learning and artificial intelligence are domains that have long been influenced by and, in turn, have influenced our understanding of how effective and efficient cognition enables animals and agents to thrive in the natural world. Systems that interact with the external world, including robots, drones, autonomous vehicles, and medical devices, depend on the ability to learn and extract meaningful sparse and compact latent signal and feature representations from large “natural domain” datasets for purposes such as image understanding, signal categorization, and real-time control. It is the case that sparse, domain-optimized representations can arise from dictionaries trained on large natural domain-specific datasets and thereby capture and encode intrinsic low-dimensional statistical structure. We give an overview of dictionary learning algorithms, both deterministic and statistical, with a discussion of the information theoretic and hierarchical Bayes concepts that motivated many of the original statistical approaches based on minimization of the entropy of factorial representations and mutual information maximization. In keeping with the machine learning emphasis of this chapter, the concept of learning a generative dictionary model as a generalization of factor analysis (FA) and independent component analysis (ICA) is discussed, as well as the concept of dictionary learning as a generalization of K-means clustering to allow hierarchical “cluster-of-clusters” concept formation and pattern recognition. We then discuss the relationship of learned dictionary-based signal processing and detection to several contemporary machine learning techniques, including nonnegative matrix factorization, clustering, discrimination, kernel methods, and deep learning, and comment on some contemporary applications of dictionary learning to machine learning problems." @default.
• W4383460730 created "2023-07-07" @default.
• W4383460730 creator A5001700017 @default.
• W4383460730 creator A5042298001 @default.
• W4383460730 creator A5056063665 @default.
• W4383460730 creator A5070446698 @default.
• W4383460730 date "2024-01-01" @default.
• W4383460730 modified "2023-10-17" @default.
• W4383460730 title "Dictionaries in machine learning" @default.
• W4383460730 cites W1890834058 @default.
• W4383460730 cites W1902027874 @default.
• W4383460730 cites W1927762903 @default.
• W4383460730 cites W1967277558 @default.
• W4383460730 cites W1976709621 @default.
• W4383460730 cites W1979191599 @default.
• W4383460730 cites W198171247 @default.
• W4383460730 cites W1982405594 @default.
• W4383460730 cites W1986736933 @default.
• W4383460730 cites W1988625484 @default.
• W4383460730 cites W1994154911 @default.
• W4383460730 cites W1995875735 @default.
• W4383460730 cites W2004465977 @default.
• W4383460730 cites W2004928594 @default.
• W4383460730 cites W2004940028 @default.
• W4383460730 cites W2008231398 @default.
• W4383460730 cites W2054256377 @default.
• W4383460730 cites W2054274648 @default.
• W4383460730 cites W2065873741 @default.
• W4383460730 cites W2067335673 @default.
• W4383460730 cites W2073237809 @default.
• W4383460730 cites W2078204800 @default.
• W4383460730 cites W2082114499 @default.
• W4383460730 cites W2085927826 @default.
• W4383460730 cites W2105464873 @default.
• W4383460730 cites W2107861471 @default.
• W4383460730 cites W2107979498 @default.
• W4383460730 cites W2111854888 @default.
• W4383460730 cites W2114508388 @default.
• W4383460730 cites W2122315118 @default.
• W4383460730 cites W2122741244 @default.
• W4383460730 cites W2127870457 @default.
• W4383460730 cites W2129131372 @default.
• W4383460730 cites W2129638195 @default.
• W4383460730 cites W2129812935 @default.
• W4383460730 cites W2132914434 @default.
• W4383460730 cites W2137234026 @default.
• W4383460730 cites W2140499889 @default.
• W4383460730 cites W2142940228 @default.
• W4383460730 cites W2143163931 @default.
• W4383460730 cites W2145096794 @default.
• W4383460730 cites W2145889472 @default.
• W4383460730 cites W2148154358 @default.
• W4383460730 cites W2152279006 @default.
• W4383460730 cites W2153663612 @default.
• W4383460730 cites W2157305458 @default.
• W4383460730 cites W2160547390 @default.
• W4383460730 cites W2161195301 @default.
• W4383460730 cites W2161765392 @default.
• W4383460730 cites W2163112044 @default.
• W4383460730 cites W2163398148 @default.
• W4383460730 cites W2166843080 @default.
• W4383460730 cites W2343141272 @default.
• W4383460730 cites W2464641472 @default.
• W4383460730 cites W2810470036 @default.
• W4383460730 cites W2919115771 @default.
• W4383460730 cites W2963030988 @default.
• W4383460730 cites W2963155023 @default.
• W4383460730 cites W2963213461 @default.
• W4383460730 cites W2963699903 @default.
• W4383460730 cites W2964012263 @default.
• W4383460730 cites W2964058567 @default.
• W4383460730 cites W2990558323 @default.
• W4383460730 cites W2991532537 @default.
• W4383460730 cites W3037761057 @default.
• W4383460730 cites W3043263114 @default.
• W4383460730 cites W3098197720 @default.
• W4383460730 cites W3098829776 @default.
• W4383460730 cites W3101749733 @default.
• W4383460730 cites W3105432754 @default.
• W4383460730 cites W4232132600 @default.
• W4383460730 cites W4237390087 @default.
• W4383460730 cites W4240035482 @default.
• W4383460730 cites W4249186015 @default.
• W4383460730 cites W4250955649 @default.
• W4383460730 cites W866015105 @default.
• W4383460730 doi "https://doi.org/10.1016/b978-0-32-391772-8.00023-5" @default.
• W4383460730 hasPublicationYear "2024" @default.
• W4383460730 type Work @default.
• W4383460730 citedByCount "0" @default.
• W4383460730 crossrefType "book-chapter" @default.
• W4383460730 hasAuthorship W4383460730A5001700017 @default.
• W4383460730 hasAuthorship W4383460730A5042298001 @default.
• W4383460730 hasAuthorship W4383460730A5056063665 @default.
• W4383460730 hasAuthorship W4383460730A5070446698 @default.
• W4383460730 hasConcept C119857082 @default.
• W4383460730 hasConcept C153180895 @default.
• W4383460730 hasConcept C154945302 @default.
• W4383460730 hasConcept C41008148 @default.

• next