Random Projection Features and Generalized Additive Models

  title={Random Projection Features and Generalized Additive Models},
  author={Subhransu Maji},
We propose to learn generalized additive models for classification which represents the classifier using a su m of piecewise linear functions and show that a recently propose d fast linear SVM training method (Pegasos) can be adapted to train such models with the same convergence rates. To be able to learn functions on combination of dimensions, we explorethe use of random projection features which learns a classifier o n data projected using an arbitrary matrix. In our experiment s we find… CONTINUE READING