Simplicial Algorithms for Minimizing Polyhedral Functions

Sold by Ingram

This product may not be approved for your region.
Paperback
  • Free Shipping

    On orders of AED 100 or more. Standard delivery within 5-15 days.
  • Free Reserve & Collect

    Reserve & Collect from Magrudy's or partner stores accross the UAE.
  • Cash On Delivery

    Pay when your order arrives.
  • Free returns

    See more about our return policy.
Polyhedral functions provide a model for an important class of problems that includes both linear programming and applications in data analysis. General methods for minimizing such functions using the polyhedral geometry explicitly are developed. Such methods approach a minimum by moving from extreme point to extreme point along descending edges and are described generically as simplicial. The best-known member of this class is the simplex method of linear programming, but simplicial methods have found important applications in discrete approximation and statistics. The general approach considered in this text, first published in 2001, has permitted the development of finite algorithms for the rank regression problem. The key ideas are those of developing a general format for specifying the polyhedral function and the application of this to derive multiplier conditions to characterize optimality. Also considered is the application of the general approach to the development of active set algorithms for polyhedral function constrained problems and associated Lagrangian forms.