000218559 001__ 218559
000218559 005__ 20180317092109.0
000218559 020__ $$a978-1-5090-1929-8
000218559 02470 $$2ISI$$a000392266500036
000218559 037__ $$aCONF
000218559 245__ $$aSource Localization on Graphs via l1 Recovery and Spectral Graph Theory
000218559 260__ $$aNew York$$bIeee$$c2016
000218559 269__ $$a2016
000218559 300__ $$a5
000218559 336__ $$aConference Papers
000218559 520__ $$aWe cast the problem of source localization on graphs as the simultaneous problem of sparse recovery and diffusion ker- nel learning. An l1 regularization term enforces the sparsity constraint while we recover the sources of diffusion from a single snapshot of the diffusion process. The diffusion ker- nel is estimated by assuming the process to be as generic as the standard heat diffusion. We show with synthetic data that we can concomitantly learn the diffusion kernel and the sources, given an estimated initialization. We validate our model with cholera mortality and atmospheric tracer diffusion data, showing also that the accuracy of the solution depends on the construction of the graph from the data points.
000218559 6531_ $$asource localization
000218559 6531_ $$agraph
000218559 6531_ $$asparsity
000218559 6531_ $$aoptimization
000218559 700__ $$0249197$$aCerqueira Gonzalez Pena, Rodrigo$$g254838
000218559 700__ $$0241065$$aBresson, Xavier$$g140163
000218559 700__ $$0240428$$aVandergheynst, Pierre$$g120906
000218559 7112_ $$a12th IEEE Image, Video, and Multidimensional Signal Processing (IVMSP) Workshop 2016$$cBordeaux, France$$dJuly 11-12, 2016
000218559 773__ $$t2016 Ieee 12Th Image, Video, And Multidimensional Signal Processing Workshop (Ivmsp)
000218559 8564_ $$uhttps://arxiv.org/abs/1603.07584$$zURL
000218559 909CO $$ooai:infoscience.tind.io:218559$$pSTI$$pconf
000218559 909C0 $$0252392$$pLTS2$$xU10380
000218559 917Z8 $$x254838
000218559 937__ $$aEPFL-CONF-218559
000218559 973__ $$aEPFL$$rREVIEWED$$sACCEPTED
000218559 980__ $$aCONF