1 / 11

Generalization Bounds for Indefinite Kernel Machines

Explore the generalization bounds for indefinite kernel machines, including positive definite and non-positive definite algorithm variants. Understand the generalization of linear separators, SVM usage, and prototype selection. Review Ali's proof and visit TTI for more insights.

whuerta
Download Presentation

Generalization Bounds for Indefinite Kernel Machines

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. Generalization Bounds for Indefinite Kernel Machines • Shai Ben-David, Ali Rahimi, Nati Srebro

  2. Xerox Phaser 8400 • About • This instance • The class • Things like this • Your location: • 1100 NE45th st, Seattle WA • Edit this page Contacts Menu

  3. … Input image Compute similarities with training images Some kind of classifier

  4. K is not... • Positive definite • A metric • A function on Rn x Rn • Differentiable

  5. Algorithm: When K is pd,use an SVM

  6. Generalization when K is pd Analyze the generalization of a class of linear separators.

  7. Algorithm: When K is not pd

  8. Algorithm: With prototype Selection

  9. Generalization Bounds whenK is indefinite Ali’s proof VisitTTI Lunchw/ Shai workshop submission workshop acceptance workshop presentation

  10. Ali’s Original Proof Idea

More Related