Can I read Stochastic Relaxation for Building Some Classes of Piecewise Linear Regression Functions on EtoBox?
Stochastic Relaxation for Building Some Classes of Piecewise Linear Regression Functions by B.A. Zalesky is a Mathematics article available to read on EtoBox.
What is Stochastic Relaxation for Building Some Classes of Piecewise Linear Regression Functions about?
The problem of building piecewise linear regression functions belonging to some classes of functions T (e.g., monotone, convex (concave), unimodal, with fixed number of local maxima (or minima), with discontinuities, etc.) by the method of stochastic relaxation (SR) is considered. Convexity of the set of admissible Solutions T is not assumed. It allows to pay attention not only to functional properties of regression functions (like continuity or "differentiability" in the sense of belonging to Sobolev or Besov spaces) but to take int o account geometrical form of functions. Numerical testing confirmes applicability of the algorithm. The results of several tests are presented in the paper.
Who reads Stochastic Relaxation for Building Some Classes of Piecewise Linear Regression Functions?
It is typically read by researchers, students, and practitioners in Mathematics.
- Author
- B.A. Zalesky
- Publisher
- Walter de Gruyter GmbH
- Published
- 2000
- Language
- EN
- Field
- Mathematics (Physical Sciences)