상세정보
학술지
Fast Training of Structured SVM Using Fixed-Threshold Sequential Minimal Optimization
- 발행일
- 200904
- 출처
- ETRI Journal, v.31 no.2, pp.121-128
- ISSN
- 1225-6463
- 출판사
- 한국전자통신연구원 (ETRI)
- 협약과제
- 08MS3700, 웹 QA 기술개발, 장명길
- 초록
- In this paper, we describe a fixed-threshold sequential minimal optimization (FSMO) for structured SVM problems. FSMO is conceptually simple, easy to implement, and faster than the standard support vector machine (SVM) training algorithms for structured SVM problems. Because FSMO uses the fact that the formulation of structured SVM has no bias (that is, the threshold b is fixed at zero), FSMO breaks down the quadratic programming (QP) problems of structured SVM into a series of smallest QP problems, each involving only one variable. By involving only one variable, FSMO is advantageous in that each QP sub-problem does not need subset selection. For the various test sets, FSMO is as accurate as an existing structured SVM implementation (SVM-Struct) but is much faster on large data sets. The training time of FSMO empirically scales between O(n) and O(n1.2), while SVM-Struct scales between O(n1.5) and O(n1.8).
- KSP 제안 키워드
- Fast training, Large data sets, Subset selection, Support VectorMachine(SVM), Training algorithms, Training time, breaks down, quadratic programming, sequential minimal optimization, structured SVM