dhb-iccopt

dhb-iccopt - Experimental Mathematics and Optimization...

Info iconThis preview shows pages 1–7. Sign up to view the full content.

View Full Document Right Arrow Icon
1 Experimental Mathematics and Optimization David H Bailey Lawrence Berkeley National Laboratory http://crd.lbl.gov/~dhbailey
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
2 Outline 1. Introduction Application: the uncertainty principle. 1. Integer relation detection The PSLQ algorithm. Applications: bifurcation constants, sculpture. 1. Sequence limit extrapolation Application: the Quinn-Rand-Strogatz constant. 1. Infinite series summation The Euler-Maclaurin formula. Euler’s transformation. Applications: Euler sums, Apery-like series, Ramanujan-like series. 1. Numerical integration Gaussian quadrature. Tanh-sinh quadrature. Applications: tetrahedral integral, box integrals, Ising integrals, spin integrals. 1. Summary
Background image of page 2
3 Methods Used in Experimental and Computer-Assisted Mathematics Computer-based symbolic computation (e.g., Mathematica, Maple, Magma ). High-precision (hundreds or thousands of digits) numerical computation. Integer relation detection algorithms (notably the PSLQ algorithm). Computer-based sequence and constant recognition facilities. Minimization and maximization schemes. Prime verification algorithms and other computational number theory methods. The Wilf-Zeilberger algorithm for proving certain infinite series identities. Formal verification methods – for instance, the effort to verify Thomas Hales’ recent proof of the Kepler conjecture using symbolic logic.
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
4 The Uncertainty Principle where Uncertainty principle from quantum mechanics: The uncertainty in position times the uncertainty in momentum must be at least some minimum value. Uncertainty principle from signal processing: A signal cannot be both time- limited and band-limited -- the dispersion of the signal times the dispersion of the Fourier transform must be at least some minimum value. The mathematical formulations of these two principles are identical:
Background image of page 4
5 A Numerical Minimization Approach to the Uncertainty Principle A formal proof of the uncertainty principle (UP) is not very difficult, but it is hardly intuitive at first reading. Here is a more computationally and intuitively appealing approach: 1. Start with a simple “tent” function. 2. Evaluate the UP product for the all perturbations of the current function on a grid with the current resolution. 3. Select the function from step 2 that has the smallest UP product. 4. Refine the grid and go to 2. 5. End when grid is sufficiently fine. Minimizing function: Gaussian function Minimum UP product: 0.25 Jonathan M. Borwein and David H. Bailey, Mathematics by Experiment , AK Peters, 2004, pg. 183-188.
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
6 The Integer Relation Problem Let (x n ) be a given vector of real numbers. An integer relation algorithm finds integers (a n ) such that (or at least within “epsilon” of zero). This can be viewed as an integer programming problem: find a set of integers (a i ) that minimizes for some M. The input (x
Background image of page 6
Image of page 7
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 11/11/2011 for the course MATH 110 taught by Professor Staff during the Winter '08 term at BYU.

Page1 / 54

dhb-iccopt - Experimental Mathematics and Optimization...

This preview shows document pages 1 - 7. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online