pen Life Lab
Back to reference library

Reference library

Strongly Typed Genetic Programming

Paper summary

Strongly typed genetic programming constrains tree construction and mutation by required return/argument types. Crossover chooses a node in the first parent and compatible-return-type candidates in the second; if no legal match exists it produces no crossover offspring or returns parents. Instantiated generic types must remain consistent after inheritance. These rules avoid many syntactically or type-invalid programs; the selected reading does not establish a quantified performance advantage.

Relevance to the project

Assistant proposal: analogous shape, named-channel and interface contracts could prevent incompatible controller exchanges. Type compatibility alone does not preserve behavior or the joint tuning of learning rules and dynamics.

Genetic representationsDigital evolution

Public author-hosted HTTP copy acquired after HTTPS certificate verification failed; verification was not disabled. Version/date differ from journal bibliographic year. Type correctness is not a guarantee of good behavior, runtime success or evolutionary benefit.

Source record

Reference ID
strongly-typed-genetic-programming-1995
Archived version
Author-hosted 34-page copy dated 20 November 2002 of the 1995 work; not asserted to be the journal version of record
Summary provenance
reference-library/papers/strongly-typed-genetic-programming-1995/metadata.json · summary
Review status
Targeted primary full-text review; not exhaustive reading or reproduction
Reading scope
Opening title/abstract and targeted author-copy PDF pp.11–13 on generic type instantiation, type-consistent mutation and crossover; final p.34 checked for identity/context. Benchmark tables and complete results not critically read; no reproduction.