Sandbox | Index | gfrlog | Contact

Light-Bot Results Gallery

This page lists all of the solutions generated by my distributed erlang genetic algorithm. The algorithm may have been run multiple times with different parameters, so each instance is listed with the relevant parameters. See the table key below for an explanation.

For each instance a link is provided to load that that instance into the viewer, where you'll be able to see not only the final solution, but also all of the intermediate solutions that were a novel improvement in the course of that algorithm.

Table Key

The first two entries (Generations and Total Commands) are results of running the algorithm, while the rest of the entries are initial parameters.

Symbol	Stands for	Explanation
G	Generations	Total number of generations that the algorithm ran for. A generation is considered to be one recombination and one mutation.
TC	Total Commands	For the best solution found, this is the number of commands executed, where commands within a function count each time they are executed
PS	Population Size	The number of organisms existing at any point while the algorithm was running
MW	Mutation Width	The number of contiguous commands changed during a mutation
RS	Recombination Segments	The number of segments the parents are divided into when creating an offspring
DR	Duplicates Replaced	If true, then when any mutation/recombination creates an organism already present in the population, it will be discarded and a completely random organism will be added instead
MI	Mutation Immunity for leader	If true, the highest ranking organism is never selected for mutation, and so the maximum value of the population never decreases
ET	Extended Time	When a solution is found, the amount of extra time allocated for looking for shorter solutions is equal to the number of generations it took to find the first solution multiplied by this number