Geometric Modeling and Imaging--New Trends (GMAI'06) A Tabu Search Meta-Heuristic for Image Semi-Supervised Classification London, England July 05-July 06 ISBN: 0-7695-2604-7
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/GMAI.2006.4
We investigate the utility of Tabu Search (TS) Meta-heuristics for semi-supervised image classification tasks. The proposed heuristic solves the integer programming Transductive Support Vector Machine (MIP-TSVM) formulation considered in [8]. Preliminary results, with a linear kernel show that our TS implementation can effectively find optimal global solutions for TSVM with relatively large problem dimensions and is competitive, in terms of generalization performance, with Transductive SVMlight package on LIBSVM benchmarks. However on Corel image database, TSVMlight demonstrates superior performance. As a result, the usefulness of such MIP-TSVM formulation may be application dependant.
Index Terms:
Tabu Search, Support Vector Machines, Mixed Integer Programming, Semi-Supervised Learning, Transductive inference, Image classification.
Citation:
Mahmoud Zennaki, Ahmed Ech-cherif, Jean Charles Lamirel, "A Tabu Search Meta-Heuristic for Image Semi-Supervised Classification," gmai, pp.239-243, Geometric Modeling and Imaging--New Trends (GMAI'06), 2006 Usage of this product signifies your acceptance of the Terms of Use. | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb