CONF Fornoni_ACML2013_2013/IDIAP Multiclass Latent Locally Linear Support Vector Machines Fornoni, Marco Caputo, Barbara Orabona, Francesco Ong, Cheng Soon Ed. Ho, Tu-Bao Ed. Latent SVM Locally Linear Support Vector Machines multiclass classification EXTERNAL http://publications.idiap.ch/attachments/papers/2013/Fornoni_ACML2013_2013.pdf PUBLIC JMLR W&CP, Volume 29: Asian Conference on Machine Learning Canberra, Australia 2013 229-244 1938-7228 http://jmlr.org/proceedings/papers/v29/ URL Kernelized Support Vector Machines (SVM) have gained the status of off-the-shelf classifiers, able to deliver state of the art performance on almost any problem. Still, their practical use is constrained by their computational and memory complexity, which grows super-linearly with the number of training samples. In order to retain the low training and testing complexity of linear classifiers and the exibility of non linear ones, a growing, promising alternative is represented by methods that learn non-linear classifiers through local combinations of linear ones. In this paper we propose a new multi class local classifier, based on a latent SVM formulation. The proposed classifier makes use of a set of linear models that are linearly combined using sample and class specific weights. Thanks to the latent formulation, the combination coe#cients are modeled as latent variables. We allow soft combinations and we provide a closed-form solution for their estimation, resulting in an effcient prediction rule. This novel formulation allows to learn in a principled way the sample specific weights and the linear classifiers, in a unique optimization problem, using a CCCP optimization procedure. Extensive experiments on ten standard UCI machine learning datasets, one large binary dataset, three character and digit recognition databases, and a visual place categorization dataset show the power of the proposed approach.