Co-evolving Fitness Predictors for Accelerating Evaluations and Reducing Sampling
نویسنده
چکیده
We introduce an estimation of distribution algorithm that co-evolves fitness predictors in order to reduce the computational cost of evolution. Fitness predictors are light objects which, given an evolving individual, heuristically approximate its true fitness. The predictors are trained by their ability to correctly differentiate between good and bad solutions using reduced computation. We apply co-evolution of fitness predictors to symbolic regression and measure its impact. Our results show that a small computational investment in co-evolving fitness predictors greatly enhances both speed and convergence of individual solutions while reducing the computational effort overall. Finally we apply fitness prediction to interactive evolution of pen stroke drawings. These results show that fitness prediction is extremely effective at modeling user preference while minimizing the sampling on the user to fewer than ten prompts.
منابع مشابه
Co-evolution of Fitness Maximizers and Fitness Predictors
We introduce an estimation of distribution algorithm (EDA) based on co-evolution of fitness maximizers and fitness predictors for improving the performance of evolutionary search when evaluations are prohibitively expensive. Fitness predictors are lightweight objects which, given an evolving individual, heuristically approximate the true fitness. The predictors are trained by their ability to c...
متن کاملSampling the Nature of a Population: Punctuated Anytime Learning for Co-evolving a Team
Evolving agents to function as cooperative members of a team is a difficult problem. In order for individuals in a team to best display heterogeneous behavior, their evolution must take place in separate populations to promote specialization. To allow the genetic algorithm to find a robust solution, the fitness evaluations must ensure that each individual is paired with team members that best r...
متن کاملRemoving Redundancy and Reducing Fitness Evaluation Costs in Genetic Programming
One of the greater issues in Genetic Programming (GP) is the computational effort required to run the evolution and discover a good solution. Phenomena such as program bloating (where genetic programs rapidly grow in size) can quickly exhaust available memory resources and slow down the evolutionary process, while the heavy cost of performing fitness evaluation can make problems which have a lo...
متن کاملRapidly Reconfigurable Field-Programmable Gate Arrays for Accelerating Fitness Evaluation in Genetic Programming
The dominant component of the computational burden of solving non-trivial problems with evolutionary algorithms is the task of measuring the fitness of each individual in each generation of the evolving population. The advent o f rapidly reconfigurable fieldprogrammable gate arrays (FPGAs) and the idea of evolvable hardware opens the possiblity of embodying each individual of the evolving popul...
متن کامل