000218019 001__ 218019
000218019 005__ 20180913063659.0
000218019 037__ $$aSTUDENT
000218019 245__ $$aStructured Auto-Encoder with application to Music Genre Recognition
000218019 269__ $$a2015
000218019 260__ $$c2015
000218019 336__ $$aStudent Projects
000218019 520__ $$aIn this work, we present a technique that learns discriminative audio features for Music Information Retrieval (MIR). The novelty of the proposed technique is to design auto-encoders that make use of data structures to learn enhanced sparse data representations. The data structure is borrowed from the Manifold Learning field, that is data are supposed to be sampled from smooth manifolds, which are here represented by graphs of proximities of the input data. As a consequence, the proposed auto-encoders finds sparse data representations that are quite robust w.r.t. perturbations. The model is formulated as a non-convex optimization problem. However, it can be decomposed into iterative sub-optimization problems that are convex and for which well-posed iterative schemes are provided in the context of the Fast Iterative Shrinkage-Thresholding (FISTA) framework. Our numerical experiments show two main results. Firstly, our graph-based auto-encoders improve the classification accuracy by 2% over the auto-encoders without graph structure for the popular GTZAN music dataset. Secondly, our model is significantly more robust as it is 8% more accurate than the standard model in the presence of 10% of perturbations.
000218019 6531_ $$aauto-encoder
000218019 6531_ $$asparse representation
000218019 6531_ $$amanifold learning
000218019 6531_ $$amusic information retrieval
000218019 6531_ $$agraph
000218019 6531_ $$anon-convex optimization
000218019 700__ $$0249515$$aDefferrard, Michaël$$g226056
000218019 720_2 $$0240428$$aVandergheynst, Pierre$$edir.$$g120906
000218019 720_2 $$0241065$$aBresson, Xavier$$edir.$$g140163
000218019 720_2 $$0247305$$aParatte, Johann$$edir.$$g174659
000218019 8564_ $$uhttps://github.com/mdeff/dlaudio$$zURL
000218019 8564_ $$s572517$$uhttps://infoscience.epfl.ch/record/218019/files/report.pdf$$yn/a$$zn/a
000218019 909C0 $$0252392$$pLTS2$$xU10380
000218019 909CO $$ooai:infoscience.tind.io:218019$$pSTI
000218019 917Z8 $$x226056
000218019 917Z8 $$x226056
000218019 917Z8 $$x226056
000218019 917Z8 $$x226056
000218019 937__ $$aEPFL-STUDENT-218019
000218019 973__ $$aEPFL
000218019 980__ $$aSTUDENT$$bMASTERS