1 / 19

Introduction to SVM

Introduction to SVM. Zhang Liliang. Outline. SVM——SVM的概念和目的 Hard Margin SVM——最原始的SVM及其对偶形式 Soft Margin SVM——引入松弛变量 Kernel——解决低维到高维的映射. The Support Vector Machine (SVM) approach. The original Support vector machines (SVMs) is a binary classification algorithm.

saber
Télécharger la présentation

Introduction to SVM

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Introduction to SVM Zhang Liliang

  2. Outline • SVM——SVM的概念和目的 • Hard Margin SVM——最原始的SVM及其对偶形式 • Soft Margin SVM——引入松弛变量 • Kernel——解决低维到高维的映射

  3. The Support Vector Machine (SVM) approach • The original Support vector machines (SVMs) is a binary classification algorithm. TARGET:Find out a linear decision surface (“hyperplane”)

  4. Case 1: Linearly separable data; “Hard-margin” linear SVM Maximize the Gap!

  5. Statement of linear SVM classifier

  6. Statement of linear SVM classifier

  7. SVM optimization problem: Primal formulation Gap(Margin): Problem Transformation: max D -> min w -> min w^2 - > min 1/2(w^2)

  8. SVM optimization problem: Dual formulation

  9. Case 2: Not linearly separable data;“Soft-margin” linear SVM

  10. Parameter C in soft-margin SVM

  11. Not linearly separable data:

  12. Kernel trick

  13. Popular kernels

  14. Conclusion • SVM:Maximize the Gap(Margin) max D -> min w -> min w^2 - > min 1/2(w^2) Hard-margin: Soft-margin: Kernel trick:

  15. Reference • http://blog.csdn.net/v_july_v/article/details/7624837(支持向量机通俗导论(理解SVM的三层境界)by July) • http://www.autonlab.org/tutorials/svm15.pdf;(来自卡内基梅隆大学carnegie mellon university(CMU)的讲解SVM的PPT) • http://www.nyuinformatics.org/downloads/supplements/SVM_Tutorial_2010/Final_WB.pdf(A Gentle Introduction to Support Vector Machines in Biomedicine)

  16. Thanks~

More Related