Hands-On Mathematics for Deep Learning
Mostprogrammersanddatascientistsstrugglewithmathematics,havingeitheroverlookedorforgottencoremathematicalconcepts.ThisbookusesPythonlibrariestohelpyouunderstandthemathrequiredtobuilddeeplearning(DL)models.You'llbeginbylearningaboutcoremathematicalandmoderncomputationaltechniquesusedtodesignandimplementDLalgorithms.Thisbookwillcoveressentialtopics,suchaslinearalgebra,eigenvaluesandeigenvectors,thesingularvaluedecompositionconcept,andgradientalgorithms,tohelpyouunderstandhowtotraindeepneuralnetworks.Laterchaptersfocusonimportantneuralnetworks,suchasthelinearneuralnetworkandmultilayerperceptrons,withaprimaryfocusonhelpingyoulearnhoweachmodelworks.Asyouadvance,youwilldelveintothemathusedforregularization,multi-layeredDL,forwardpropagation,optimization,andbackpropagationtechniquestounderstandwhatittakestobuildfull-fledgedDLmodels.Finally,you’llexploreCNN,recurrentneuralnetwork(RNN),andGANmodelsandtheirapplication.Bytheendofthisbook,you'llhavebuiltastrongfoundationinneuralnetworksandDLmathematicalconcepts,whichwillhelpyoutoconfidentlyresearchandbuildcustommodelsinDL.
·7.4萬字