liveBook · Manning

Selective pressure

As explained in Optimization Algorithms: AI techniques for design, planning, and control problems

The relative likelihood that the fittest individual is selected compared with the average likelihood of selection across the population.

Selective pressure measures how much more likely the fittest individual is to be selected than an average population member. It governs the trade-off between exploiting strong solutions and preserving diversity for exploration. section 7.3.3

Higher selective pressure produces faster but riskier convergence:

  • Population diversity decreases more quickly, increasing the risk of premature convergence on a suboptimal solution. section 7.3.3
  • Elitism has the highest selective pressure because it preserves top-performing individuals and reduces succeeding-generation diversity. section 7.3.3
  • Fitness-proportionate selection assigns each individual a selection probability proportional to its fitness and samples from that distribution. section 7.3.3

Ticket pricing example

Five candidate solutions receive fitness-proportionate probabilities: x4, the best candidate, receives 0.39; x3 receives 0.27; and lower-ranked candidates receive less. Elitism selects x4 and x3 as parents for offspring.

How Optimization Algorithms: AI techniques for design, planning, and control problems covers it

How deep it goes
Defines the measure, explains diversity and convergence effects, and illustrates fitness-proportionate probabilities and elitism.
Examples
Uses a ticket-pricing population, ranked binary candidates, roulette-wheel probabilities, cumulative probabilities, and selection-method figures.
What it assumes
Assumes familiarity with genetic-algorithm populations, individuals, fitness, parent selection, crossover, and mutation.
What you can do after
Evaluate selection operators by their pressure, diversity loss, convergence risk, and exploration consequences.
How it connects
Connects genetic algorithms with parent selection, elitism, fitness-proportionate selection, rank-based selection, and tournament selection.
Cover of Optimization Algorithms: AI techniques for design, planning, and control problems
liveBookmanning.com
Read Optimization Algorithms: AI techniques for design, planning, and control problems on liveBook Get the book at manning.com

This explainer was generated by Manning's knowledge graph from the text of Optimization Algorithms: AI techniques for design, planning, and control problems. The section links point to the parts of the book it draws on.