Semi-supervised Penalized Output Kernel Regression for Link Prediction

Link prediction is addressed as an output kernel learning task through semi-supervised Output Kernel Regression. Working in the framework of RKHS theory with vector-valued functions, we establish a new representer theorem devoted to semi-supervised least square regression. We then apply it to get a new model (POKR: Penalized Output Kernel Regression) and show its relevance using numerical experiments on artificial networks and two real applications using a very low percentage of labeled data in a transductive setting.

[1]  É. Senkene,et al.  Hilbert spaces of operator-valued functions , 1973 .

[2]  John D. Lafferty,et al.  Diffusion Kernels on Graphs and Other Discrete Input Spaces , 2002, ICML.

[3]  B. Snel,et al.  Function prediction and protein networks. , 2003, Current opinion in cell biology.

[4]  Ben Taskar,et al.  Link Prediction in Relational Data , 2003, NIPS.

[5]  Bernhard Schölkopf,et al.  Learning with Local and Global Consistency , 2003, NIPS.

[6]  Kiyoshi Asai,et al.  The em Algorithm for Kernel Matrix Completion with Auxiliary Data , 2003, J. Mach. Learn. Res..

[7]  Yoshihiro Yamanishi,et al.  Protein network inference from multiple genomic data: a supervised approach , 2004, ISMB/ECCB.

[8]  Gal Chechik,et al.  Euclidean Embedding of Co-occurrence Data , 2004, J. Mach. Learn. Res..

[9]  Mikhail Belkin,et al.  Semi-Supervised Learning on Riemannian Manifolds , 2004, Machine Learning.

[10]  William Stafford Noble,et al.  Kernel methods for predicting protein-protein interactions , 2005, ISMB.

[11]  Jason Weston,et al.  A general regression technique for learning transductions , 2005, ICML '05.

[12]  Tsuyoshi Kato,et al.  Selective integration of multiple biological data for supervised network inference , 2005, Bioinform..

[13]  Charles A. Micchelli,et al.  On Learning Vector-Valued Functions , 2005, Neural Computation.

[14]  Mikhail Belkin,et al.  Manifold Regularization: A Geometric Framework for Learning from Labeled and Unlabeled Examples , 2006, J. Mach. Learn. Res..

[15]  Pierre Geurts,et al.  Kernelizing the output of tree-based methods , 2006, ICML '06.

[16]  C. Micchelli,et al.  Universal Multi-Task Kernels , 2008, J. Mach. Learn. Res..

[17]  Pierre Geurts,et al.  Gradient boosting for kernelized output spaces , 2007, ICML '07.

[18]  Pierre Geurts,et al.  Inferring biological networks with output kernel trees , 2007, BMC Bioinformatics.

[19]  Jon M. Kleinberg,et al.  The link-prediction problem for social networks , 2007, J. Assoc. Inf. Sci. Technol..

[20]  Jean-Philippe Vert,et al.  Supervised reconstruction of biological networks with local models , 2007, ISMB/ECCB.

[21]  Charles A. Micchelli,et al.  Universal Multi-Task Kernels , 2008, J. Mach. Learn. Res..

[22]  Yoshihiro Yamanishi,et al.  propagation: A fast semisupervised learning algorithm for link prediction , 2009 .

[23]  Thomas L. Griffiths,et al.  Nonparametric Latent Feature Models for Link Prediction , 2009, NIPS.

[24]  Charles A. Micchelli,et al.  When is there a representer theorem? Vector versus matrix regularizers , 2008, J. Mach. Learn. Res..

[25]  Stéphane Canu,et al.  Nonlinear functional regression: a functional RKHS approach , 2010, AISTATS.