1
Lecture 4-6: Dimension Reduction
One common method is to assume smooth density functions in empty spaces, e.g.
Lecture notes Stat 231-CS276A S.C. Zhu
Dimension reduction techniques
The other method is to reduce the dimension of the feature space, for example by projecting
a feature vector to a lower dimensional space.
Common techniques for dimension reduction:
1. Principle component analysis (PCA)
2. Fisher linear discriminant analysis
3. Independent component analysis (ICA)
4 Multi
dimensional scaling (MDS)
Lecture notes Stat 231-CS276A, S.C. Zhu
4
.
Multi
-
dimensional scaling (MDS)
5. Over-complete bases coding
6. Transformed component analysis (TCA)
