Learning Bounds for Support Vector Machines with Learned Kernels

By Nati Srebro, Shai Ben-David. 2006

99 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Interpretable Distribution Features with Maximum Testing Power (2016, background): “The VC property will be a direct consequence of the VC indices of finite-dimensional function classes and preservation theorems (see Lemma 7); for a nice example application see Srebro and Ben-David (2006) (Section 5) who study the pseudo-dimension of (x,y) 7→ k(x,y) kernel classes, for different Ga”
Regularization Techniques for Learning with Matrices (2009, methodology): “We cannot compare our bound directly to the bound in Srebro and Ben-David [2006] as they do not work with Rademacher complexities.”
Multi-task and Lifelong Learning of Kernels (2015, methodology): “By using covering numbers, Srebro et al in [4] have shown generalization bounds with additive dependence on the pseudodimension of the kernel family.”
SPARSITY IN MULTIPLE KERNEL LEARNING (2010, background): “…to as the multiple kernel learning, has been studied recently by Bousquet and Herrmann (2003), Crammer, Keshet and Singer (2003), Lanckriet et al. (2004), Micchelli and Pontil (2005), Lin and Zhang (2006), Srebro and Ben-David (2006), Bach (2008) and Koltchinskii and Yuan (2008) among others.”
L2 Regularization for Learning Kernels (2009, methodology): “…of learning kernels has been investigated in a number of recent publications including (Lanckriet et al., 2004; Micchelli & Pontil, 2005; Argyriou et al., 2005; Ar-gyriou et al., 2006; Srebro & Ben-David, 2006; Ong et al., 2005; Lewis et al., 2006; Zien & Ong, 2007; Jebara, 2004; Bach, 2008).”
Minimax Optimal Rates of Estimation in High Dimensional Additive Models: Universal Phase Transition (2015, background): “These type of problems have been studied previously by Bousquet et al. (2003), Cramer et al. (2003), Lanckriet et al. (2004), Micchelli and Pontil (2005), Srebro and Ben-David (2006), Bach (2008), and Suzuki and Sugiyama (2013) among others.”
Generalization bounds for learning the kernel (2009, methodology): “However, considering that our main target is to estimate generalization bounds and excess misclassification errors, the direct approach [25] would result in quite loose generalization bounds.”