Timeline for Can the reproduction function of an evolutionary algorithm consider gene strength?
Current License: CC BY-SA 3.0
11 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 28, 2016 at 22:20 | vote | accept | PeterTheLobster | ||
| S Apr 28, 2016 at 19:56 | history | suggested | Brad Larson |
When I migrated this over from Stack Overflow, the tags got lost. This aligns them with the proper tags on CS.
|
|
| Apr 28, 2016 at 17:01 | review | Suggested edits | |||
| S Apr 28, 2016 at 19:56 | |||||
| Apr 28, 2016 at 17:00 | history | migrated | from stackoverflow.com (revisions) | ||
| Apr 28, 2016 at 9:25 | answer | added | manlio | timeline score: 2 | |
| Apr 24, 2016 at 21:37 | comment | added | dfri | Sounds related to the common GA concept of elitism; since your best incumbent solution is your best scoring individual, you can make sure to keep one or a few copies of this individual in each new generation (elitism), or use a few copies as base for "good individuals" with only few random mutations without crossover (bringing these to next gen.). This will implicitly bring good genes over to your scoring functions, as the genes of the best individual (via elitism) will, probabilistically after say tournament selection, contribute to new individuals via e.g. mutations/crossover etc. | |
| Apr 24, 2016 at 21:36 | comment | added | PeterTheLobster | @Ray Great thanks! I was just wondering wether that was against a certain standard or definition. I personally believe that you should do whatever works best, but this is for an assignement so I wanted make sure. | |
| Apr 24, 2016 at 21:30 | comment | added | Ray | Can you? Of course you can. There are no rules here. You can add the "gene strength" to the fitness function. You can do anything you want as long as it makes the solution converge faster. | |
| Apr 24, 2016 at 21:23 | comment | added | PeterTheLobster | @Ray What I mean by that is that for example if I could measure how much a certain gene contributed to the solution attempt can I give the gene a higher chance to be passed on to the next generation or should child genes be mixed randomly? | |
| Apr 24, 2016 at 21:14 | comment | added | Ray | I'm not sure what you mean by "gene strength". However, if you can measure "gene strength" then you can certainly use it in your reproduction probability. | |
| Apr 24, 2016 at 19:17 | history | asked | PeterTheLobster | CC BY-SA 3.0 |