Optimize by Generation (AGA) (AI Studio Core)

Parameters

• max number of new attributesMax number of attributes to generate for an individual in one generation.
• limit max total number of attributesIndicates if the total number of attributes in all generations should be limited.
• max total number of attributesMax total number of attributes in all generations.
• use local random seedIndicates if a local random seed should be used.
• local random seedSpecifies the local random seed
• maximal fitnessThe optimization will stop if the fitness reaches the defined maximum.
• population sizeNumber of individuals per generation.
• maximum number of generationsNumber of generations after which to terminate the algorithm.
• use plusGenerate sums.
• use diffGenerate differences.
• use multGenerate products.
• use divGenerate quotients.
• reciprocal valueGenerate reciprocal values.
• use early stoppingEnables early stopping. If unchecked, always the maximum number of generations is performed.
• generations without improvalStop criterion: Stop after n generations without improval of the performance.
• tournament sizeThe fraction of the current population which should be used as tournament members (only tournament selection).
• start temperatureThe scaling temperature (only Boltzmann selection).
• dynamic selection pressureIf set to true the selection pressure is increased to maximum during the complete optimization run (only Boltzmann and tournament selection).
• keep best individualIf set to true, the best individual of each generations is guaranteed to be selected for the next generation (elitist selection).
• p initializeInitial probability for an attribute to be switched on.
• p crossoverProbability for an individual to be selected for crossover.
• crossover typeType of the crossover.
• p generateProbability for an individual to be selected for generation.
• use heuristic mutation probabilityIf checked the probability for mutations will be chosen as 1/number of attributes.
• p mutationProbability for mutation.
• use square rootsGenerate square root values.
• use power functionsGenerate the power of one attribute and another.
• use sinGenerate sinus.
• use cosGenerate cosinus.
• use tanGenerate tangens.
• use atanGenerate arc tangens.
• use expGenerate exponential functions.
• use logGenerate logarithmic functions.
• use absolute valuesGenerate absolute values.
• use minGenerate minimum values.
• use maxGenerate maximum values.
• use sgnGenerate signum values.
• use floor ceil functionsGenerate floor, ceil, and rounded values.
• restrictive selectionUse restrictive generator selection (faster).
• remove uselessRemove useless attributes.
• remove equivalentRemove equivalent attributes.
• equivalence samplesCheck this number of samples to prove equivalency.
• equivalence epsilonConsider two attributes equivalent if their difference is not bigger than epsilon.
• equivalence use statisticsRecalculates attribute statistics before equivalence check.
• search fourier peaksUse this number of highest frequency peaks for sinus generation.
• attributes per peakUse this number of additional peaks for each found peak.
• epsilonUse this range for additional peaks for each found peak.
• adaption typeUse this adaption type for additional peaks.