Next:
Theory
Contact:
Peter Mills
Adaptive Gaussian filtering is a simple and powerful implementation of
variable bandwidth kernel estimators for classification, PDF estimation
and interpolation. The library incorporates several innovations to produce
one of the fastest and most accurate supervised classification algorithms
in the world. These include:
- matching kernel width to sample density quickly and accurately via the properties of the exponential function
- restricting calculations to a set of k-nearest-neighbours found in n log k time with a binary tree
- generating a pre-trained model by searching for the class-borders with guaranteed, superlinear convergence
- extrapolating the conditional probabilities to provide solid knowledge of estimate accuracy
Author: Peter Mills
Next: Theory