NSGA-II

As explained in Evolutionary Deep Learning: Genetic algorithms and neural networks

NSGA-II is a multi-objective evolutionary algorithm that uses elitist non-dominated sorting and crowding to preserve diverse high-quality solutions.

NSGA-II is a multi-objective evolutionary algorithm and selection operator that balances fitness optimization with long-term population diversity. It is especially useful when evolution runs long enough for populations to specialize or become stuck. section 3.4

Use the selection experiment to inspect how NSGA-II behaves:

• Set the population size to 1000 and run 15 generations to make differences between selection methods visible. section 3.4
• Inspect the final evolution output and compare histogram shapes as indicators of population diversity. section 3.4
• Contrast NSGA-II with best, worst, tournament, and SPEA2 selection when studying specialization and Pareto-front distribution. section 3.4

Figure 3.7 A comparison of selection operators on population diversity

Compare the population distributions produced by different selection operators, including NSGA-II. section 3.4

How Evolutionary Deep Learning: Genetic algorithms and neural networks covers it

How deep it goes

Defines elitist non-dominated sorting and crowding, then emphasizes observed diversity rather than deriving the algorithm formally.

Examples

Compares selection-operator histograms using populations of 1000 individuals evolved for 15 generations.

What you can do after

Apply NSGA-II in an evolutionary experiment and inspect whether its population remains diverse over longer runs.

How it connects

Connects NSGA-II to multi-objective optimization, Pareto fronts, selection operators, and population diversity.

liveBookmanning.com

Read Evolutionary Deep Learning: Genetic algorithms and neural networks on liveBook Get the book at manning.com

Related concepts

• genetic algorithmkind of · A genetic algorithm is an evolutionary optimization method that improves a population of candidate solutions through fitness-based selection and variation, often including recombination and mutation.
• diversityproduces · Diversity is the degree of variation among members of a set or population in relevant characteristics, perspectives, behaviors, or outcomes.
• pymoouses · A Python framework for multi-objective optimization that provides evolutionary algorithms, metaheuristics, visualization tools, and decision-making capabilities.

This explainer was generated by Manning's knowledge graph from the text of Evolutionary Deep Learning: Genetic algorithms and neural networks. The section links point to the parts of the book it draws on.