Population-based incremental learning: Difference between revisions

Revision as of 20:45, 26 February 2019

Latest revision as of 08:36, 1 December 2020

(2 intermediate revisions by 2 users not shown)

Line 13:

| publisher = Carnegie Mellon University

| place = Pittsburgh, PA

| url = http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.61.8554

| issue = CMU–CS–94–163

| citeseerx = 10.1.1.61.8554

}}</ref><ref name="Baluja2">{{Citation

| last1 = Baluja | first1 = Shumeet

| last2 = Caruana | first2 = Rich

Line 22:

| publisher = Morgan Kaufmann Publishers

| pages = 38–46

| urlciteseerx = http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.44.5424

}}</ref><ref name="Baluja3">{{Citation

| last1 = Baluja | first1 = Shumeet

| title = An Empirical Comparison of Seven Iterative and Evolutionary Function Optimization Heuristics

Line 29:

| publisher =

| pages =

| urlciteseerx = http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.43.1108

}}</ref>

== Algorithm ==

Line 41:

# Update population genotype (probability vector) based on fittest individual.

# Mutate.

# Repeat steps 1-41–4

== Source code ==

This is a part of source code implemented in [[Java (programming language)|Java]]. In the paper, learnRate = 0.1, negLearnRate = 0.075, mutProb = 0.02, and mutShift = 0.05 is used. N = 100 and ITER_COUNT = 1000 is enough for a small problem.

<sourcesyntaxhighlight lang="java">

public void optimize() {

final int totalBits = getTotalBits();

Line 111:

}

}

</syntaxhighlight>

</source>

==See also==