Context-aware crossover

HeuristicLab-Trac-Bot commented 12 years ago

Issue migrated from trac ticket # 1693

milestone: HeuristicLab 3.3.7 | component: Problems.DataAnalysis.Symbolic | priority: medium | resolution: done

2011-12-01 10:11:40: @foolnotion created the issue

Given a randomly selected subtree from one parent, context-aware crossover will always find the best location to place the subtree in the other parent.

In their paper ([1]), Majeed and Ryan suggest that using standard crossover in the initial part of the run and then switching to context-aware crossover yields the best performance. They show that, across a range of standard GP benchmark problems, context-aware crossover produces a higher best fitness as well as a higher mean fitness, and even manages to solve the 11-bit multiplexer problem without ADFs. At the same time, the individuals produced this way are much smaller than standard GP, and far fewer individual evaluations are required, so GP achieves a higher fitness by evaluating fewer and smaller individuals.

The authors hypothesize that the order in which crossover operators are applied (for instance the standard and the context-aware ones) and the time of their application can have a big impact on the experiments outcome.

[1] Using Context-aware Crossover to Improve the Performance of GP, Proceeding, GECCO '06, Proceedings of the 8th annual conference on Genetic and evolutionary computation, ACM New York, NY, USA ©2006.

HeuristicLab-Trac-Bot commented 12 years ago

2011-12-01 17:16:33: @foolnotion changed status from new to assigned

HeuristicLab-Trac-Bot commented 12 years ago

2011-12-01 17:16:33: @foolnotion commented

r7109: Implemented the context-aware crossover.

HeuristicLab-Trac-Bot commented 12 years ago

2011-12-01 17:16:50: @foolnotion changed status from assigned to reviewing

HeuristicLab-Trac-Bot commented 12 years ago

2011-12-01 17:16:50: @foolnotion changed owner from @foolnotion to @mkommend

HeuristicLab-Trac-Bot commented 12 years ago

2011-12-01 17:20:30: @foolnotion commented

r7110: Submitted another variant of the context-aware crossover, called the DeterministicBestCrossover. This one selects a random crossover point and finds the best node from the other parent to be inserted there.

HeuristicLab-Trac-Bot commented 12 years ago

2011-12-01 17:20:30: @foolnotion

HeuristicLab-Trac-Bot commented 12 years ago

2012-02-20 13:12:50: @mkommend commented

Reviewing comments:

selection of cutpoints

reorder criteria for cutpoint selection

swap method

do not set the operator name in the default ctor as this is done by the ItemAttribute automatically

improve descriptions

HeuristicLab-Trac-Bot commented 12 years ago

2012-02-20 13:12:50: @mkommend changed owner from @mkommend to @foolnotion

HeuristicLab-Trac-Bot commented 12 years ago

2012-02-20 13:12:50: @mkommend changed status from reviewing to assigned

HeuristicLab-Trac-Bot commented 12 years ago

2012-02-21 11:01:28: @foolnotion commented

r7495: Simplified selection of cutpoints and reordered criteria for cutpoint selection

HeuristicLab-Trac-Bot commented 12 years ago

2012-02-21 14:40:22: @foolnotion changed owner from @foolnotion to @mkommend

HeuristicLab-Trac-Bot commented 12 years ago

2012-02-21 14:40:22: @foolnotion changed status from assigned to reviewing

HeuristicLab-Trac-Bot commented 12 years ago

2012-02-21 14:41:48: @mkommend changed milestone from HeuristicLab 3.3.x Backlog to HeuristicLab 3.3.7

HeuristicLab-Trac-Bot commented 12 years ago

2012-02-23 16:06:51: @mkommend changed owner from @mkommend to @foolnotion

HeuristicLab-Trac-Bot commented 12 years ago

2012-02-23 16:06:51: @mkommend changed status from reviewing to readytorelease

HeuristicLab-Trac-Bot commented 12 years ago

2012-07-08 04:35:53: @gkronber removed resolution

HeuristicLab-Trac-Bot commented 12 years ago

heal-research / HeuristicLab