Second-order optimization of variable projection SVM models and road abnormality detection
Organizations: Faculty of Mathematics and Computer Science, UniDistance Suisse, Brig, Switzerland · CEMSE Division, KAUST, Thuwal, Saudi Arabia · Eötvös Loránd University, Faculty of Informatics, Department of Numerical Analysis
Abstract
We introduce a novel second-order optimization framework for minimizing so-called variable projection functionals. We demonstrate that the proposed framework is especially usefulfor the training of variable projection based kernel methods. In particular, the problem of efficiently training variable projection support vector machines (VP-SVMs) is considered. We show the effectiveness of the proposed training methodology in a real-world application, namely we demonstrate how second-order trust region algorithms can be used to train VPSVM models to recognize road surface abnormalities based on 1D signals obtained from a tire sensor.
Figures & tables
| Alg. | It | Avg Acc. | Min Acc. | Max Acc. | Time (s) | |
| TR | 15 | 95.5% | 92.2% | 9.16e-04 | ||
| 25 | - | 5.39 | ||||
| GD | 25 | 72.3% | 68.0% | 77.9% | 1.57e-02 | |
| 125 | 83.9% | 79.6% | 88.5% | 7.52e-03 | 1.68 | |
| 250 | 88.2% | 85.4% | 93.3% | 4.82e-03 | 3.39 | |
| 500 | 91.3% | 89.3% | 95.2% | 2.10e-03 | 6.78 |