Name
Abstract | ||
|---|---|---|
OneClassMaxMinOver (OMMO) is a simple incre- mental algorithm for one-class support vector classifi- cation. We propose several enhancements and heuris- tics for improving model selection, including the adap- tation of well-known techniques such as kernel caching and the evaluation of the feasibility gap. Furthermore, we provide a framework for optimising grid search based model selection that compromises of preinitial- isation, cache reuse, and optimal path selection. Finally, we derive simple heuristics for choosing the optimal grid search path based on common benchmark datasets. In total, the proposed modifications improve the runtime of model selection significantly while they are still simple and adaptable to a wide range of incre- mental support vector algorithms. |
Year | DOI | Venue |
|---|---|---|
2008 | 10.1109/ICPR.2008.4761775 | Tampa, FL |
Keywords | Field | DocType |
minimax techniques,pattern classification,support vector machines,MaxMinOver-based training,benchmark datasets,cache reuse,fast model selection,incremental algorithm,incremental support vector algorithms,kernel caching,optimal grid search path,optimal path selection,support vector classification,support vector machines | Structured support vector machine,Data mining,Computer science,Heuristics,Artificial intelligence,Least squares support vector machine,Pattern recognition,Support vector machine,Model selection,Relevance vector machine,Kernel method,Sequential minimal optimization,Machine learning | Conference |
ISSN | ISBN | Citations |
1051-4651 E-ISBN : 978-1-4244-2175-6 | 978-1-4244-2175-6 | 1 |
PageRank | References | Authors |
0.36 | 6 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Fabian Timm | 1 | 11 | 1.99 |
| Sascha Klement | 2 | 24 | 3.26 |
| Thomas Martinetz | 3 | 1462 | 231.48 |