More generality in efficient multiple kernel learning

By Manik Varma, Bodla Rakesh Babu. International Conference on Machine Learning, 2009

467 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Signal Processing

View Paper: https://www.semanticscholar.org/paper/aba8a31a1df26cb3fa5c4d8d701b45a3c1324652

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

In Signal Processing

Citations in the corpus

l p -Norm Multiple Kernel Learning (2011, background): “Although finding non-linear kernel mixtures (Gönen and Alpaydin, 2008; Varma and Babu, 2009) generally results in non-convex optimization problems, Cortes et al.”
A Survey on Multi-view Learning (2013, background): “57% UCI2 LP-SVM MKL GMKL (Varma and Babu, 2009) Ionosphere 93.”
Feature Selection as a One-Player Game (2010, methodology): “Other approaches use non-linear hypotheses (Shen et al., 2008; Varma & Babu, 2009) or random forests (Breiman et al., 1984; Rogers & Gunn, 2005), to compute feature scores.”
Cross-Domain Object Recognition Via Input-Output Kernel Analysis (2013, methodology): “Therefor e, we propose learning the output kernel matrix directly with a non-parametric formulation [17] and learning the input ker nel matrix with a parametric form similar to Multiple Kernel Learning (MKL) [26]–[28].”
Deep multilayer multiple kernel learning (2015, methodology): “During the last decade, MKL has been actively investigated, in which lots of algorithms and techniques were proposed in order to improve the efficiency of the MKL method [8, 13, 19, 22, 24, 27, 28, 30].”
Multiple Kernel Learning Algorithms (2011, unknown): “The gradient with respect to the kernel weights is calculated as: Varma and Ray (2007); Varma and Babu (2009) perform gender identiﬁcation experiments on a face image data set by combining kernels calculated on each individual feature, and hence, for kernels whose η m goes to 0 perform feature…”
Optimization for machine learning (2010, methodology): “It is being applied to fusing heterogeneous data (Lanckriet et al., 2004a; Sonnenburg et al., 2006b; Zien and Ong, 2007; Rakotomamonjy et al., 2008; Varma and Babu, 2009), to understand the learned kernel classifier (Sonnenburg et al.”
Action bank: A high-level representation of activity in video (2012, unknown): “3 Varma and Babu [35] 85.”
Methodologies for Cross-Domain Data Fusion: An Overview (2015, methodology): “Nonlinear combination methods [63] use nonlinear functions of kernels, namely, multiplication, power, and exponentiation.”
A review on machine learning principles for multi-view biological data integration (2016, unknown): “GMKL [163] Generalized MKL based on gra-”
Scalable hierarchical multitask learning algorithms for conversion optimization in display advertising (2014, methodology): “Our work builds on well known multiple kernel learning [19] and collaborative filtering techniques namely that of effectively imposing a hyperprior on the regularization term.”
Algorithms for Learning Kernels Based on Centered Alignment (2012, background): “However, the performance reported for these other families does not seem to be consistently superior to that of the uniform combination either.”
Learning Non-Linear Combinations of Kernels (2009, background): “Non-linear combinations of kernels have been recently cons idered by [23].”
Boosted local structured HOG-LBP for object localization (2011, methodology): “MKL has great advantage to handle multiple, heterogeneous data sources and has been widely applied in the problem of feature selection and combination [23, 24].”
Insights from Classifying Visual Concepts with Multiple Kernel Learning (2011, methodology): “Alternatives use more general combinations of kernels such as products with kernel widths as weighting parameters [30, 31].”
Prediction of the Leadership Style of an Emergent Leader Using Audio and Visual Nonverbal Features (2017, unknown): “The performance of LMKL was compared with single kernel SVM and another popular MKL method called Generalized Multiple Kernel Learning (GMKL) [72], [73].”
Multiple Kernel Fuzzy Clustering (2012, background): “Varma and Babu studied superlinear combinations of kernels [32], and Gonen et al. studied local combinations of kernels [33].”
Detection Evolution with Multi-order Contextual Co-occurrence (2013, methodology): “Many popular feature selection algorithms have been proposed, such as Boosting [11, 12] or Multiple Kernel Learning [31, 30].”
Kernel Methods on the Riemannian Manifold of Symmetric Positive Definite Matrices (2013, methodology): “As a feature selection method, MKL has proven more effective than conventional feature selection methods such as wrappers, filters and boosting [21].”