170 likes | 245 Views
Age-Based Population Dynamics in Evolutionary Algorithms. Lisa Guntly. Motivation. Parameter specification complicates EAs Optimal parameter values can change during a run Age is an important factor in Biology. The Importance of Age.
E N D
Age-Based Population Dynamics in Evolutionary Algorithms Lisa Guntly
Motivation • Parameter specification complicates EAs • Optimal parameter values can change during a run • Age is an important factor in Biology
The Importance of Age • Age significantly impacts survival in natural populations
F i S S i AGE F B Methods • Survival chance (Si) of an individual is based on age and fitness • Main Equation Fitness of i = Best Fitness
S R ( AGE ) AGE A Survival Chance from Age • Age is tracked by individual, and is incremented every generation • Two equations explored for SAGE • Equation 1 (ABPS-AutoEA1): linear decrease 1 = - Rate of decrease from age
N AG S AGE Survival Chance from Age (cont’d) • Equation 2 (ABPS-AutoEA2): more dynamic Number of individuals in the same age group AGE 1 = - - 2 P 2 G Population size Number of generations the EA will run
N AGE = - - AG S 1 AGE 2 P 2 G Survival Chance from Age (cont’d) • Effects of • More individuals of the same age will decrease their survival chance • Age will decrease survival chance relative to the maximum age (G) NAG Si
Experimental Setup • Testing done on TSP (size 20/40/80) • Offspring size is constant • Compared to a manually tuned EA • Examine effects of • Initial population size • Offspring size • Tracked population statistics • Size • Average age
Performance Results - TSP size 20 Average over 30 runs ABPS-AutoEA1 - ABPS-AutoEA2 -
Performance Results - TSP size 40 Average over 30 runs ABPS-AutoEA1 - ABPS-AutoEA2 -
Initial Population Size Effect 3 different runs
Tracking Population Size and Average Age Same single run
F F F i i W S S AGE AGE F F F B B W Equation with Fitness Scaling • Attempt to fix the lack of selection pressure from fitness • New Main Equation Fitness of i - Fitness Scaling S S = = i i - Best Fitness Worst Fitness
Initial Performance Analysis from Fitness Scaling Equation Average over 30 runs using
Initial Performance Analysis from Fitness Scaling Equation (cont’d) • Elitism improved performance slightly • Roulette wheel (fitness proportional) parent selection improved performance on a larger TSPs but performed worse on smaller TSPs • Independence from initial population size was maintained • Adjustment of population size during the run was improved
Future Work • Further exploration of fitness scaling methods • Test on additional problems • Compare to other dynamic population sizing schemes • Implement age-based offspring sizing