03509nam a22005535i 4500001001800000003000900018005001700027007001500044008004100059020003700100024003500137050001500172072001600187072002300203072001500226100008400241245018900325250001800514264006900532300006300601336002600664337002600690338003600716347002400752490007500776505036100851520070001212650003801912650002801950650003401978650004902012650002902061650003602090650007502126700009002201700008102291700008902372700008502461700008802546710003402634773002602668776003602694776003602730776003602766830007502802856004602877912001502923999001702938978-3-031-38133-1DE-He21320251030110812.0cr nn 008mamaa230901s2024    sz |    s    |||| 0|eng d  a97830313813319978-3-031-38133-17 a10.1007/978-3-031-38133-12doi 4aTK7895.E42 7aUKM2bicssc 7aCOM0920002bisacsh 7aUKM2thema1 aAlsuhli, Ghada.eauthor.4aut4http://id.loc.gov/vocabulary/relators/aut96585610aNumber Systems for Deep Neural Network Architecturesh[electronic resource] /cby Ghada Alsuhli, Vasilis Sakellariou, Hani Saleh, Mahmoud Al-Qutayri, Baker Mohammad, Thanos Stouraitis.  a1st ed. 2024. 1aCham :bSpringer Nature Switzerland :bImprint: Springer,c2024.  aXI, 94 p. 30 illus., 28 illus. in color.bonline resource.  atextbtxt2rdacontent  acomputerbc2rdamedia  aonline resourcebcr2rdacarrier  atext filebPDF2rda1 aSynthesis Lectures on Engineering, Science, and Technology,x2690-03270 aIntroduction -- Conventional number systems -- DNN architectures based on Logarithmic Number System (LNS) -- DNN architectures based on Residue Number System (RNS) -- DNN architectures based on Block Floating Point (BFP) number system -- DNN architectures based on Dynamic Fixed Point (DFXP) number system -- DNN architectures based on Posit number system.  aThis book provides readers a comprehensive introduction to alternative number systems for more efficient representations of Deep Neural Network (DNN) data. Various number systems (conventional/unconventional) exploited for DNNs are discussed, including Floating Point (FP), Fixed Point (FXP), Logarithmic Number System (LNS), Residue Number System (RNS), Block Floating Point Number System (BFP), Dynamic Fixed-Point Number System (DFXP) and Posit Number System (PNS). The authors explore the impact of these number systems on the performance and hardware design of DNNs, highlighting the challenges associated with each number system and various solutions that are proposed for addressing them. 0aEmbedded computer systems.964625 0aMicroprocessors.959300 0aComputer architecture.965857 0aNeural networks (Computer science) .96423114aEmbedded Systems.96585824aProcessor Architectures.96585924aMathematical Models of Cognitive Processes and Neural Networks.9658601 aSakellariou, Vasilis.eauthor.4aut4http://id.loc.gov/vocabulary/relators/aut9658611 aSaleh, Hani.eauthor.4aut4http://id.loc.gov/vocabulary/relators/aut9658621 aAl-Qutayri, Mahmoud.eauthor.4aut4http://id.loc.gov/vocabulary/relators/aut9658631 aMohammad, Baker.eauthor.4aut4http://id.loc.gov/vocabulary/relators/aut9658641 aStouraitis, Thanos.eauthor.4aut4http://id.loc.gov/vocabulary/relators/aut9658652 aSpringerLink (Online service)0 tSpringer Nature eBook08iPrinted edition:z978303138132408iPrinted edition:z978303138134808iPrinted edition:z9783031381355 0aSynthesis Lectures on Engineering, Science, and Technology,x2690-032740uhttps://doi.org/10.1007/978-3-031-38133-1  aZDB-2-SXSC  c56259d56259