Genetic Algorithms and Benevolent Viruses

Seed idea. Sparked by Children of Time — the nanovirus that accelerates evolution.

The hook

A “benevolent virus” spreads useful traits horizontally across a population, not just parent-to-child. In GAs this is a real technique (Virus-Evolutionary GA / Kanoh et al.): a virus carries a good schema and infects others to speed convergence.

Threads to grow

• Standard GA vs GA + horizontal-transfer “virus” operator — convergence comparison.
• What makes a virus “benevolent” vs disruptive? When does horizontal transfer hurt?
• Real-world parallel: horizontal gene transfer in microbes.
• Synthesis with Neural Networks and Brain Scaling: neuroevolution + a “neurovirus” that spreads beneficial weight-patterns.

Side Project / Prompt Seed

If I hand this file to Claude: let’s build something. Start by digging up relevant research, then prototype a proof of concept in code.

Premise: a “benevolent virus” that accelerates evolution by spreading good traits horizontally across a population.

What to build (POC): a plain GA on a toy problem (OneMax, knapsack, or a small maze-solver). Add a “virus” operator: extract a high-fitness sub-solution (schema) from a strong individual and inject it into others. Run with/without the virus; plot convergence speed and final fitness. Tune infection rate and see when it helps vs. when it collapses diversity.

Research to pull first:

• Virus-Evolutionary GA (Kanoh et al., 1990s) — the direct prior art.
• Horizontal gene transfer in microbial evolution (the biological analogue).
• Memetic algorithms / island models — other “beyond vertical inheritance” GA variants.

Stretch: make the host a small NN (neuroevolution) and the virus spread weight-patterns — the literal “neurovirus.” Bridges to Neural Networks and Brain Scaling and Uplift and Engineered Evolution.