A meta-learning approach to automatic kernel selection for support vector machines

By Shawkat Ali, K. Smith‐Miles. 2006

151 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Marco F. Huber
2. Marc-André Zöller
3. Wei-Wei Tu
4. Quanming Yao
5. Yang Yu
6. Qiang Yang
7. Yi-Qi Hu
8. Ricardo Bastos Cavalcante Prudêncio

In Computer Vision and Pattern Recognition

1. Quanming Yao
2. Hugo Jair Escalante
3. Isabelle M Guyon
4. Marc-André Zöller
5. André Carlos Ponce de Leon Ferreira de Carvalho
6. Yang Yu
7. Qiang Yang
8. Marco F. Huber

Citations in the corpus

• Benchmark and Survey of Automated Machine Learning Frameworks (2019, methodology): “Metalearning can be used as an additional criterion for selecting promising candidate configurations based on the predicted performance (e.g., Alia & Smith-Miles, 2006; Wistuba et al., 2015b; Nargesian et al., 2017) or ranking of the models (e.g., Sohn, 1999; Gama & Brazdil, 2000).”
• Survey on Automated Machine Learning (2019, positive): “sually selecting the conguration with the highest expected improvement. Meta-learning can be used to lter empirically bad performing candidate congurations based on the predicted performance e.g., (Alia and Smith-Miles, 2006; Wistuba et al., 2015a; Nargesian et al., 2017) or ranking of the models, e”
• Support Vector Machine Versus Random Forest for Remote Sensing Image Classification: A Meta-Analysis and Systematic Review (2020, methodology): “Accordingly, innovative approaches, including automatic kernel selection [75], [79], [81], and multiple kernel learning [82], were proposed to address this problem.”
• Combining meta-learning and search techniques to select parameters for support vector machines (2012, methodology): “An alternative approach to SVM parameter selection is the use of meta-learning, which treats the SVM parameter selection as a supervised learning task [1,5].”
• Taking Human out of Learning Applications: A Survey on Automated Machine Learning (2018, methodology): “For this purpose, metaknowledge indicating the empirically good configurations are extracted, and the meta-learner take the characteristics of learning problem as input and predict promising configurations, such as kernel [112], adaptive network architectures [113], [114].”
• An Experimental Study of the Combination of Meta-Learning with Particle Swarm Algorithms for SVM Parameter Selection (2012, methodology): “ML is an alternative that has been studied in recent years to SVM parameter selection [1][15][11][16][17][5].”