Topological Complexity of Algorithms

Abstract

The research was directed toward understanding methods for calculating complexity of algorithms in non-classical sense. Topological complexity measures the necessity of branching in any possible algorithm for solving given problem. Particular problems which I considered are solving polynomial equations of special type (i.e. with several apriori vanishing coefficients). The difficulty is in the lack at the present time detailed information on geometry and topology of discriminants of the space of polynomial equations of the type in question. I was able to find the degrees of these discriminants and resolved questions of irreducibility. Topological complexity was found completely for trinomial equations. Major steps was made toward investigating of algorithms for solving systems of 2 polynomial equations with 2 unknowns. I essentially calculated the fundamental group of the space of such system obtaining close relations to the Artin's braid group. Preliminary investigation was made in the meaning of topological complexity of solving differential equations. (JHD)

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 12, 1990
Accession Number
ADA220274

Entities

People

  • A. Libgober

Organizations

  • University of Illinois at Chicago

Tags

DTIC Thesaurus Topics

  • Abstracts
  • Algebra
  • Algorithms
  • Coefficients
  • Differential Equations
  • Equations
  • Geometry
  • Mathematics
  • Military Research
  • Polynomials
  • Security
  • Topology
  • Triangles

Readers

  • Graph Algorithms and Convex Optimization.
  • Linear Algebra
  • Theoretical Analysis.

Technology Areas

  • Space