Genetic Algorithm

Evolution as an algorithm

Genetic algorithms were created by John Holland in the 1970s, borrowing Darwinian evolution to search for good solutions. A candidate is encoded as a “chromosome”; a population of them evolves through a loop — score each on a fitness function, select the fitter ones as parents, recombine them with crossover, and apply a little random mutation to keep diversity. Repeat for generations.

They are genuinely powerful on hard problems: crew rostering, factory scheduling, engineering design. NASA’s ST5 mission even flew a GA-evolved antenna whose odd, organic shape outperformed conventional designs — and when the orbit changed, engineers simply adjusted the fitness function and re-evolved a new one in under a month.

Fitness is whatever you define

That last detail is the whole point: a genetic algorithm optimises exactly the objective you write into its fitness function, and only that. Change the objective and the “best” solution changes with it. The algorithm has no goals of its own — it climbs the landscape you hand it.

Here, the fitness rewards an even spread, varied digits, a balanced mix and a typical sum. It is real evolutionary computation optimising a real objective — but that objective is a look: “scattered, balanced, unremarkable in sum.” There is no gradient toward winning to climb, because a fair draw is random and memoryless; the landscape for “will this ticket win” is flat.

How we borrow its shape

The Genetic method powers every step — the starting population, selection, crossover and mutation — from genuine quantum entropy, evolving candidate tickets toward that aesthetic and returning the fittest. The randomness driving the evolution is real; the objective is simply how the numbers look.

• —Rewards a broad, even spread across the whole range
• —Favours varied digits and a balanced odd/even mix
• —Keeps the total near the typical, central sum band

Sources & further reading

1. John Holland — Genetic Algorithms (foundational overview) — The field’s originator on population, fitness-based selection, crossover and mutation.
2. Wikipedia — Evolved antenna — NASA ST5’s GA-designed antenna — the first evolved hardware flown in space.
3. World Lotteries Association — random chance is the essence of the lottery — Industry statement that draws are random and independent.
4. The Math Doctors — the gambler’s fallacy — Why memoryless draws mean no number is ever “due.”