Title: CS558 Project
1 CS558 Project
- Local SVM Classification based on triangulation
(on the plane) - Glenn Fung
2Outline of Talk
- Classification problem on the plane
- All of the recommended stages were applied
- Sampling
- Ordering
- Clustering
- Triangulation
- Interpolation (Classification)
- SVM Support vector Machines
- Optimization Number of training points
increased - Evaluation
- Checkerboard dataset
- Spiral dataset
3Classification Problem in
4SAMPLING 1000 randomly sampled points
5ORDERING Clustering
- A Fuzzy-logic based clustering algorithm was
used. - 32 cluster centers were obtained
6ORDERING Delaunay Triangulation
- Algorithms to triangulate and to get the
Delaunay triangulation from HWKs 3 and 4 were
used. - Given a point,the random point approach is used
to localize the triangle that contains it.
7Interpolation SVM
- SVM Support Vector Machine Classifiers
-
- A different nonlinear Classifier is used for
each triangle - The triangle structure is efficiently used for
both training and testing phases and for defining
a simple and fast nonlinear classifier.
8What is a Support Vector Machine?
- An optimally defined surface
- Typically nonlinear in the input space
- Linear in a higher dimensional space
- Implicitly defined by a kernel function
9What are Support Vector Machines Used For?
- Classification
- Regression Data Fitting
- Supervised Unsupervised Learning
(Will concentrate on classification)
10Support Vector MachinesMaximizing the Margin
between Bounding Planes
A
A-
11The Nonlinear Classifier
- Where K is a nonlinear kernel, e.g.
12Reduced Support Vector Machine AlgorithmNonlinear
Separating Surface
13How to Choose in RSVM?
14(No Transcript)
15Obtained Bizarre Checkerboard
16Optimization More sampled points Training
parameters adjusted
17Result Improved Checkerboard
18Nonlinear PSVM Spiral Dataset94 Red Dots 94
White Dots
19NextBascom Hill
20Some Questions
- Would it work for BW pictures (regression
instead of classification?