Efficient training of graph-regularized multitask SVMs Conference Paper
| Authors: | Widmer, C.; Kloft, M.; Görnitz, N.; Ratsch, G. |
| Title: | Efficient training of graph-regularized multitask SVMs |
| Conference Title: | European Conference, ECML PKDD 2012 |
| Abstract: | We present an optimization framework for graph-regularized multi-task SVMs based on the primal formulation of the problem. Previous approaches employ a so-called multi-task kernel (MTK) and thus are inapplicable when the numbers of training examples n is large (typically n < 20,000, even for just a few tasks). In this paper, we present a primal optimization criterion, allowing for general loss functions, and derive its dual representation. Building on the work of Hsieh et al. [1,2], we derive an algorithm for optimizing the large-margin objective and prove its convergence. Our computational experiments show a speedup of up to three orders of magnitude over LibSVM and SVMLight for several standard benchmarks as well as challenging data sets from the application domain of computational biology. Combining our optimization methodology with the COFFIN large-scale learning framework [3], we are able to train a multi-task SVM using over 1,000,000 training points stemming from 4 different tasks. An efficient C++ implementation of our algorithm is being made publicly available as a part of the SHOGUN machine learning toolbox [4]. © 2012 Springer-Verlag. |
| Keywords: | computational biology; algorithms; benchmarking; bioinformatics; support vector machines; data sets; optimization; application domains; computational experiment; dual representation; learning frameworks; loss functions; multi-task kernels; optimization criteria; optimization framework; optimization methodology; three orders of magnitude; training example |
| Journal Title | Lecture Notes in Computer Science |
| Volume: | 7523 |
| Issue: | Pt. 1 |
| Conference Dates: | 2012 Sep 24-28 |
| Conference Location: | Bristol, United Kingdom |
| ISBN: | 0302-9743 |
| Publisher: | Springer |
| Date Published: | 2012-01-01 |
| Start Page: | 633 |
| End Page: | 647 |
| Language: | English |
| DOI: | 10.1007/978-3-642-33460-3_46 |
| PROVIDER: | scopus |
| DOI/URL: | |
| Notes: | Chapter in "Machine Learning and Knowledge Discovery in Databases" (ISBN: 978-3-642-33459-7) -- "Conference code: 93009" - "Export Date: 2 November 2012" - "Sponsors: University of Bristol; Departments of Computer Science and Engineering Mathematics; Faculty of Engineering; PASCAL2 Network of Excellence; EternalS FP7 Coordination Action" - "Source: Scopus" |
Altmetric
Citation Impact
BMJ Impact Analytics
Related MSK Work