A survey of the state of the art in learning the kernels

Abbasnejad, M. Ehsan; Ramachandram, Dhanesh; Rajeswari, M.

doi:10.1007/s10115-011-0404-6

Cited by 39 publications

(16 citation statements)

References 75 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…e interested reader may refer to the recent survey on kernel learning by Abbasnejad et al [2012]. ese approaches are thus powerful but limited to the transductive setting: the resulting kernel is difficult to use on new data.…”

Section: Related Topicsmentioning

confidence: 99%

Metric Learning

Bellet

Habrard²,

Sebban³

2015

Synthesis Lectures on Artificial Intelligence and Machine Learn

108

103

View full text Add to dashboard Cite

International audienceSimilarity between objects plays an important role in both human cognitive processes and artificial systems for recognition and categorization. How to appropriately measure such similarities for a given task is crucial to the performance of many machine learning, pattern recognition and data mining methods. This book is devoted to metric learning, a set of techniques to automatically learn similarity and distance functions from data that has attracted a lot of interest in machine learning and related fields in the past ten years. In this book, we provide a thorough review of the metric learning literature that covers algorithms, theory and applications for both numerical and structured data. We first introduce relevant definitions and classic metric functions, as well as examples of their use in machine learning and data mining. We then review a wide range of metric learning algorithms, starting with the simple setting of linear distance and similarity learning. We show how one may scale-up these methods to very large amounts of training data. To go beyond the linear case, we discuss methods that learn nonlinear metrics or multiple linear metrics throughout the feature space, and review methods for more complex settings such as multi-task and semi-supervised learning. Although most of the existing work has focused on numerical data, we cover the literature on metric learning for structured data like strings, trees, graphs and time series. In the more technical part of the book, we present some recent statistical frameworks for analyzing the generalization performance in metric learning and derive results for some of the algorithms presented earlier. Finally, we illustrate the relevance of metric learning in real-world problems through a series of successful applications to computer vision, bioinformatics and information retrieval

show abstract

Section: Related Topicsmentioning

confidence: 99%

Metric Learning

Bellet

Habrard²,

Sebban³

2015

Synthesis Lectures on Artificial Intelligence and Machine Learn

108

103

View full text Add to dashboard Cite

show abstract

“…It is another way to obtain the kernel that the weighting coefficients of the basic kernels are optimized [28][29][30][31].…”

Section: Mk-svm With the Unweighted Sum Of Baisc Kernelsmentioning

confidence: 99%

“…Support vector machine (SVM) is a state-of-art machine learning approach [26,27]. Multiple kernel support vector machine (MK-SVM) is a popular topic in kernel methods [28][29][30][31]42,44,45]. It aims at learning the optimal kernel function by optimizing the combination of multiple heterogeneous basic kernels [31,45], multiple basic kernels with different feature subsets [28] or hyperparameters [44].…”

Section: Introductionmentioning

confidence: 99%

“…Multiple kernel support vector machine (MK-SVM) is a popular topic in kernel methods [28][29][30][31]42,44,45]. It aims at learning the optimal kernel function by optimizing the combination of multiple heterogeneous basic kernels [31,45], multiple basic kernels with different feature subsets [28] or hyperparameters [44]. MK-SVM has been solved by an optimization method such as semi-infinite linear program, quadraticallyconstrained quadratic program, gradient-based approaches and block coordinator descent approach [28][29][30][31]42,44,45].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Distributed customer behavior prediction using multiplex data: A collaborative MK-SVM approach

Chen

Fan

2012

Knowledge-Based Systems

View full text Add to dashboard Cite

“…As for the codebook, we set the number of clusters to be 1024 in k-means, and randomly select 8 * 10 4 features from the whole training feature set to generate codebook for each data set. Then the ScSPM [42] method is used for coding, in which the spatial block number on each level of the pyramid is set as [1,2,4], the weight for features on each level is set as . We use the matlab toolbox of BPDN-homotopy provided on the webpage http://users.ece.gatech.edu/%7Esasif/homotopy/ (The same is also provided on http://www.eecs.berkeley.edu/∼yang/software/l1benchmark/).…”

Section: A Experimental Setupmentioning

confidence: 99%

Cooperative Sparse Representation in Two Opposite Directions for Semi-Supervised Image Annotation

Zhao

Glotin

Xie

et al. 2012

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

Recent studies have shown that sparse representation (SR) can deal well with many computer vision problems, and its kernel version owns more powerful classification capability. In this paper, we address the application of a cooperative sparse representation (Co-SR) in semi-supervised image annotation which can grow the amount of labeled images for further use in training image classifiers. Provided a set of labeled (training) images and a set of unlabeled (test) images, the usual SR method, which we call forward SR, is to represent each unlabeled image with several labeled ones, and then to annotate the unlabeled image according to the annotations of these labeled ones. However, to the best of our knowledge, the SR method in an opposite direction, which we call backward SR, to represent each labeled image with several unlabeled images, and then to annotate any unlabeled image according to the annotations of the labeled images which the unlabeled image is selected by the backward SR to represent, has not been addressed by researchers. In this paper, we explore how much the backward SR can contribute to image annotation, and be complementary to the forward SR. The co-training, which has been proved to be a semi-supervised method improving each other only if two classifiers are relatively independent, is then adopted to testify this complementary nature between two SRs in opposite directions. Finally, the co-training of two SRs in kernel space builds a cooperative kernel sparse representation (Co-KSR) method for image annotation. LGC (Local and Global Consistency) and GFHF (Gaussian Fields and Harmonic Functions). Comparative experiments with a non-sparse solution are also performed to show that the sparsity plays an important role in the cooperation of image representations in two opposite directions. Our work will extend the application of sparse representation in image annotation and retrieval.

show abstract

A survey of the state of the art in learning the kernels

Cited by 39 publications

References 75 publications

Metric Learning

Metric Learning

Distributed customer behavior prediction using multiplex data: A collaborative MK-SVM approach

Cooperative Sparse Representation in Two Opposite Directions for Semi-Supervised Image Annotation

Contact Info

Product

Resources

About