# Changes between Version 5 and Version 6 of Documentation/Reference/Genetic Programming Crossovers

Ignore:
Timestamp:
01/31/12 11:00:40 (9 years ago)
Comment:

--

Unmodified
Removed
Modified
• ## Documentation/Reference/Genetic Programming Crossovers

 v5 - It selects a compatible subtree in parent1 (subject to size limits and grammar constraints) - It performs the swap and returns the (now different) parent0 The selection (step 2) above is influenced by a configurable bias towards internal or leaf nodes, given by the !InternalCrossoverPointProbability parameter2. The selection (step 2) above is influenced by a configurable bias towards internal or leaf nodes, given by the !InternalCrossoverPointProbability parameter. === 2 Probabilistic functional crossover === Given a randomly selected crossover point from parent1, context-aware crossover will always find the best location to place the subtree into parent0. Because of the initial random step, the context-aware crossover does guarantee that the resulting offspring will be better than its parents. It makes the best choice given a specific piece of genetic material. The complexity depends on the number of nodes in the first parent, since the algorithm will perform a swap followed by a tree evaluation for every available position, so that it can find out the best insertion point. This means ''n'' evaluations for ''n'' available positions, giving an O(n) complexity. A discussion of this crossover can be found in [2]. A discussion of this crossover can be found in [''2'']. === 5 Deterministic-best crossover ===