Skip to content

Feature Request: Diversity preservation mechanisms #352

Description

@zs001-agi

Feature Request: Diversity Preservation

First, thanks for the great library! PyGAD is an excellent GA framework.

I've been working on a related project (asi-evolve) that focuses specifically on diversity preservation in evolutionary algorithms. One challenge I've noticed in standard GA implementations is premature convergence — the population converges to a local optimum before exploring the search space adequately.

Proposed Feature

Add built-in diversity preservation mechanisms:

  1. Fitness sharing — reduce fitness of similar individuals
  2. Niching — maintain subpopulations in different regions
  3. Crowding — replace similar parents with offspring

Reference Implementation

I've implemented these in asi-evolve under the EvolutionaryAlgorithm class. The key method is _calculate_diversity() which tracks population diversity and auto-adjusts mutation rates when diversity drops.

Would this be a welcome addition to PyGAD? I'd be happy to contribute a PR.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions