Free cookie consent management tool by TermsFeed Policy Generator

Opened 14 years ago

Closed 14 years ago

Last modified 13 years ago

#976 closed enhancement (done)

Implement Algorithms: Offspring Selection Genetic Algorithm

Reported by: abeham Owned by: abeham
Priority: high Milestone: HeuristicLab 3.3.0
Component: Algorithms.OffspringSelectionGeneticAlgorithm Version: 3.3
Keywords: Cc:

Description

We need an offspring selection GA.

Change History (22)

comment:1 Changed 14 years ago by abeham

  • Status changed from new to assigned

comment:2 Changed 14 years ago by abeham

r3378

  • added draft of OSGA with mainloop adapted from GA

comment:3 Changed 14 years ago by abeham

r3379

  • added first working version of OSGA

comment:4 Changed 14 years ago by abeham

r3404

  • added gender specific selection

comment:5 Changed 14 years ago by abeham

r3413

  • fixed GS, added ConditionalSelector, removed parameter, removed os plugin, set linear comparison factor adjustment as default,...

comment:6 Changed 14 years ago by abeham

r3418

  • added StochasticMultiBranch and MultiCrossover / PermutationMultiCrossover

comment:7 Changed 14 years ago by abeham

r3419

  • fixed oversight in MultiPermutationCrossover

comment:8 Changed 14 years ago by abeham

r3425

  • refactored StochasticMultiOperators

comment:9 Changed 14 years ago by abeham

r3426

  • made comparison factor modifier optional
  • added evaluated solutions to show in the results (but decoupled from ToString changes)

comment:10 Changed 14 years ago by abeham

r3427

  • moved some variable initialization and assignment to a value parameter

comment:11 Changed 14 years ago by abeham

r3429

  • added a crude offspring selection island ga

comment:12 Changed 14 years ago by abeham

r3445

  • updated multicrossover, multimutation

comment:13 Changed 14 years ago by abeham

r3446

  • updated os island ga

comment:14 Changed 14 years ago by gkronber

OSGA and SASEGASA should hold only the minimal required number of individuals at any time. This is especially important for solution encodings that have a large memory footpring (e.g. SymbolicExpressionTree).

In particular special care should be taken that the number of individuals in the set of successful individuals + the number of unsuccessful individuals + the number of currently selected individuals + the number of individuals in the current population is not significantly larger than 2x population size.

comment:15 Changed 14 years ago by abeham

r3510

  • added a parameter to set the selected parents

comment:16 Changed 14 years ago by abeham

r3611 Changed Island OSGA

  • removed mainloop
  • added mainoperator

comment:17 Changed 14 years ago by abeham

r3699

  • fixed replacement in OSGA, Island OSGA, and SASEGASA

comment:18 Changed 14 years ago by abeham

r3713

  • changed elites replacement in all offspring selection algorithms

comment:19 Changed 14 years ago by abeham

r3769

  • fixed a bug in the OffspringSelector

comment:20 Changed 14 years ago by abeham

  • Resolution set to fixed
  • Status changed from assigned to closed

Offspring selection algorithms have been implemented for 3.3

Version 0, edited 14 years ago by abeham (next)

comment:21 Changed 13 years ago by swagner

  • Milestone changed from Iteration 4 to Current

Milestone Iteration 4 deleted

comment:11 Changed 13 years ago by swagner

  • Milestone changed from Current to HeuristicLab 3.3.0

Milestone Current deleted

Note: See TracTickets for help on using tickets.