Abstract | ||
---|---|---|
The fusion of multimodal sensor informa- tion often requires learning decision rules from sam- ples of high-dimensional data. Each data dimension may only be weakly informative for the detection prob- lem of interest. Also, it is not known a priori which components combine to form a lower-dimensional fea- ture space that is most informative. To learn both the combination of dimensions and the decision rule specified in the reduced-dimensional space together, we jointly optimize the linear dimensionality reduction and margin-based supervised classification problems, repre- senting dimensionality reduction by matrices on the Stiefel manifold. We describe how the learning proce- dure and resulting decision rule can be implemented in parallel, serial, and tree-structured fusion networks. |
Year | Venue | Keywords |
---|---|---|
2009 | Seattle, WA | matrix algebra,pattern classification,sensor fusion,stiefel manifold,decision rule learning,dimensionality-reduced classifiers,information fusion,linear dimensionality reduction,margin-based supervised classification,multimodal sensor information,sensor network,supervised classification,feature space,decision rule,optimization,asynchronous transfer mode,high dimensional data,support vector machines,manifolds,data mining,tree structure |
Field | DocType | ISBN |
Decision rule,Feature vector,Semi-supervised learning,Dimensionality reduction,Pattern recognition,Computer science,Support vector machine,Stiefel manifold,Curse of dimensionality,Sensor fusion,Artificial intelligence,Machine learning | Conference | 978-0-9824-4380-4 |
Citations | PageRank | References |
3 | 0.39 | 12 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Kush R. Varshney | 1 | 3 | 0.39 |
Alan S. Willsky | 2 | 7466 | 847.01 |