Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Bounding the errors for convex dynamics on one or more polytopes.

Charles Tresser1

  • 1IBM T. J. Watson Research Center, P.O. Box 218, Route 134, Kitchawan Road, Yorktown Heights, New York 10598, USA.

Chaos (Woodbury, N.Y.)
|October 2, 2007
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Bell inequality violations under reasonable and under weak hypotheses.

Physical review letters·2013
Same author

Introduction: pattern formation at the turn of the millennium.

Chaos (Woodbury, N.Y.)·2004
Same author

Convex dynamics: unavoidable difficulties in bounding some greedy algorithms.

Chaos (Woodbury, N.Y.)·2004
Same author

On the geometry of master-slave synchronization.

Chaos (Woodbury, N.Y.)·2003
Same author

Master-slave synchronization and the Lorenz equations.

Chaos (Woodbury, N.Y.)·2003
Same author

Piecewise linear models for the quasiperiodic transition to chaos.

Chaos (Woodbury, N.Y.)·1996

This study analyzes a greedy algorithm for approximating polytope sequences, focusing on Euclidean error norms. Researchers proved the existence of invariant sets for bounded polytope families, challenging prior conjectures on error saturation.

Area of Science:

  • Computational Geometry
  • Optimization Algorithms
  • Dynamical Systems

Background:

  • The greedy algorithm approximates input sequences within polytopes using vertex sequences.
  • Error bounds were previously established for single polytopes using invariant regions in affine spaces.
  • General cases with multiple polytopes showed no invariant affine space regions.

Purpose of the Study:

  • To investigate the existence of invariant sets for a greedy algorithm approximating sequences in families of polytopes.
  • To analyze the behavior of cumulative errors in dynamical systems derived from this algorithm.
  • To challenge existing conjectures regarding error saturation in polytope approximation.

Main Methods:

  • Interpreting the greedy algorithm as a time-dependent dynamical system in vector and affine spaces.

Related Experiment Videos

  • Proving the existence of large convex invariant sets in the vector space for bounded polytope families with finite outgoing normals.
  • Demonstrating error set saturation issues for specific single polytopes (quadrilaterals, pentagons), contradicting previous conjectures.

    • Main Results:

    • Existence of arbitrary large convex invariant sets in the vector space is proven under specific conditions (bounded sizes, finite normals).
    • Error saturation in finite steps is shown to occur even for single quadrilaterals and axially symmetric pentagons.
    • This disproves a former conjecture and suggests a potential pathway for proving bounded errors in general polytope families.

    Conclusions:

    • The study establishes conditions for invariant set existence in vector spaces for greedy polytope approximation.
    • A key conjecture regarding error saturation is disproven, highlighting complexities in multi-polytope systems.
    • Findings suggest that bounding errors in general polytope families may be achievable with further theoretical development.