[非负矩阵分解及在人脸识别的应用.ppt

Non-Negative Matrix Factorization（NMF） Reportor: MaPeng 作者的相关信息 Daniel D. Lee, Ph.D. Associate ProfessorDept. of Electrical and Systems EngineeringDept. of Bioengineering (Secondary)GRASP (General Robotics, Automation, Sensing, Perception) Lab 203B Moore/6314University of Pennsylvania200 S. 33rd StreetPhiladelphia, PA 19104215-898-8112215-573-2068(FAX) Email: ddlee@ /~ddlee/ H. Sebastian Seung Professor of Computational Neuroscience, MITInvestigator, Howard Hughes Medical Institute MIT, 46-506543 Vassar St.Cambridge, MA 02139voice: 617-252-1693seung@ Administrative assistant:Amy Dunnvoice: 617-452-2694fax: 617-452-2913adunn@ /people/seung/ Problem Statement Given a set of images: Create a set of basis images that can be linearly combined to create new images Find the set of weights to reproduce every input image from the basis images Dimension reduction Mainly Discuss PCA NMF LNMF FNMF WNMF What don’t we like about PCA? PCA involves adding up some basis images then subtracting others Basis images aren’t physically intuitive Subtracting doesn’t make sense in context of some applications How do you subtract a face? What does subtraction mean in the context of document classification? back Non-negative Matrix Factorization Like PCA, except the coefficients in the linear combination cannot be negative Non-negative matrix factorization (NMF) NMF is based on Gradient Descent NMF: V?WH s.t. Wi,d,Hd,j?0 Let C be a given cost function, then update the parameters according to: The idea behind multiplicative updates The NMF decomposition is not unique NMF Basis Imagesnmf_basis Only allowing adding of basis images makes intuitive sense Has physical analogue in neurons Forcing the reconstruction coefficients to be positive leads to nice basis images To reconstruct images, all you can do is add in more basis images This leads to basis images that represent parts Faces Training set: 2429 examples First 25 examples shown at right Set consists of 19x19 ce