AIDA-CMK: Multi-Algorithm Optimization Kernel Applied to Analog IC Sizing by Ricardo Lourenço Nuno Lourenço & Nuno Horta

Author:Ricardo Lourenço, Nuno Lourenço & Nuno Horta
Language: eng
Format: epub
Publisher: Springer International Publishing, Cham

3.3 Nondominated Sorting Genetic Algorithm II (NSGA-II)

NSGA-II [7] kernel is an evolutionary optimization scheme. The principle of evolutionary computation is to mimic natural evolution. The genetic algorithm starts by generating an initial population of chromosomes, the initial parents. This first population must offer a wide diversity of genetic materials. The gene pool should be as large as possible so that any solution of the search space can be engendered but generally, the initial population is generated randomly. Then, the genetic algorithm evolves the solutions by applying the genetic operators and then selecting the next parents. The process is repeated until the convergence or ending criterion is reached. The algorithm is stopped when the population converges toward the optimal solution.

New solution vectors are obtained from the current population by the application of the genetic operators of mutation and crossover. Crossover uses genes from two population elements to generate the new elements, combining randomly selected sets of information from each of the parents into the children. Mutation is a random change in individual’s genetic information in order to escape from local minima; the mutation operator introduces new information in the chromosome whereas the crossover selected the best pieces of the information present in the population genetic information.

Each chromosome has an associated value corresponding to the fitness of the solution it represents. The fitness should correspond to an evaluation of how good the candidate solution is. Selection compares each individual in the population by using a fitness function. The new individuals’ fitness is evaluated and, then, they are ranked together with the parents. The fittest individuals are selected as the new parents, and the less fit discarded.

In the particular case of the NSGA-II, the algorithm pseudocode is shown in Algorithm 3.1. NSGA-II uses Pareto dominance concepts to sort the multi-objective solutions.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.