19 Questions 65 Answers 0 Followers
Questions related from Stephen Chen
My search problem is essentially sphere with a noisy evaluation. This noise is causing CMA-ES to converge at a ridiculously slow rate. Is there any way to force CMA-ES to converge in an...
25 October 2022 5,190 3 View
I am looking to develop an overview/survey of specific experimental techniques and papers in which exploration is defined, measured, and analyzed as part of heuristic search (preferably for...
30 October 2018 5,381 12 View
The choice of something to ruin can be an implicit choice as to what should be preserved. A heuristic for preservation can thus lead to a heuristic for ruin. I've had what I think is a very...
08 March 2017 366 1 View
Most search techniques for continuous search spaces use some type of gradient -- one solution is better than another, so solutions move in that direction. A specific example is the attraction...
13 January 2015 7,474 14 View
We are all familiar with the proliferation of metaphor-based metaheuristics. From the stalwarts of Simulated Annealing, Genetic Algorithms, Ant Colony Optimization, and Particle Swarm...
30 November 2014 8,174 11 View
The BBOB (2009) benchmark functions include functions that are unimodal, have "adequate global structure", and have "weak global structure". From the problem definitions, these search space...
05 November 2014 8,033 5 View
If we have that information, then we can do some interesting things to exploit that knowledge to improve the performance of some search techniques. Our clustering based methods to measure the...
21 May 2014 6,105 10 View
Let us start with a simple 1+lambda-ES. The possible search points will be centered around the current position. I'm going to call this centering exploitation. In general, given a point, we can...
06 May 2014 8,149 5 View
In general, xnew = x1 + F*(x2-x3) where x2 and x3 are randomly selected. If x2 and x3 are ordered so that the difference vector is applied on a downward slope (for a minimization problem),...
13 December 2013 5,416 4 View
I recently read an excellent paper detailing the problems with metaphors in the design of metaheuristics:...
03 November 2013 2,347 5 View
When we draw diagrams of PSO vectors, we usually draw three points -- the current position, the pbest position and the lbest position. We then add three vectors, velocity, attraction to pbest,...
26 August 2013 300 38 View
This is a follow up thread to https://www.researchgate.net/post/What_is_the_optimal_number_of_restarts_for_a_fixed_number_of_function_evaluations? The goal is to develop general guidelines that...
16 August 2013 8,326 6 View
First, the size of the search space grows exponentially. For each dimension d, there are two extremes, so there are 2^d "corners" of the search space. For d = 1000, there are 2^1000 ~= 10^9...
22 July 2013 6,016 10 View
At CEC2013, a presenter said that Storn and Price recommended a population size of 10 times the number of dimensions -- e.g. population size = 100 for a ten dimensional problem. However, the only...
05 July 2013 6,688 31 View
Many benchmarks for optimization are based on a fixed number of function evaluations -- e.g. 5000*dimension. Many search techniques (e.g. PSO, DE, etc) can be tuned to converge quickly or more...
07 May 2013 2,177 18 View
If the algorithm is very greedy (e.g. local search), then restarts of that algorithm might resemble a memetic algorithm. Let's consider a memetic algorithm with a very expensive local...
04 April 2013 6,978 5 View
Something that covers PSO, DE, ES, EDA, etc.
20 November 2012 9,342 17 View
Large amounts of research focus on the balance between Exploration and Exploitation. However, this omits the power of Selection which can turn any metaheuristic into a hill climber by rejecting...
01 January 1970 4,011 6 View
A recurring theme in metaheuristics research is to consider the balance between Exploration and Exploitation. An often forgotten area of research is the effect of Selection on Search/Exploration....
01 January 1970 5,314 5 View
