Theory and Practice of Support Vector Machines Optimization

Shai Shalev-Shwartz*, Nathan Srebo

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

2 Scopus citations
Original languageAmerican English
Title of host publicationAutomatic Speech and Speaker Recognition
Subtitle of host publicationLarge Margin and Kernel Methods
PublisherJohn Wiley & Sons, Ltd
Number of pages16
ISBN (Print)9780470696835
StatePublished - 14 Jan 2009
Externally publishedYes


  • Bias term incorporation
  • Binary classification and traditional SVM
  • Cost-sensitive multiclass categorization
  • Dual decomposition methods
  • Gradient-based methods and loss functions
  • SVM training and linear prediction models
  • Sequence prediction and cost-sensitive multi-class categorization
  • Stochastic gradient descent (SGD) approach and training SVMs
  • Support Vector Machines (SVMs) optimization
  • approximation error with low-norm linear predictor

Cite this