Sphere
Problem definition
Objective function
f = @(x) x(:,1).^2 + x(:,2).^2
Optimization settings
o = struct% initializing struct
o.lb = -2% lower bounds
o.ub = 2% upper bounds
Graphic representation
[x,y] = meshgrid(o.lb:0.5:o.ub) surf(x,y,f([x(:),y(:)]))
Problem properties
convexity | smoothness | minimum |
1 | ∞ | f(0,0) = 0 |
Optimization example with fminsearch
Optimization
rng(0)% for tractability
x0 = [2,2]% initial guess
[xmin,fmin,info] = fminsearch(f,x0,o)% running minimization
Animation
rng(0) x0 = [2,2] [~,~,info] = fminsearch(f,x0,o) info.sol = [0 0 0]% solution
info.animate = true% plot animation
info.animfreq = 6% frame frequency
optimview('fminsearch',info)
References
[1] M. A. Schumer, K. Steiglitz, "Adaptive Step Size Random
Search", IEEE Transactions on Automatic Control. vol. 13, no. 3, pp. 270-276, 1968