Co-evolutionary Probabilistic Structured Grammatical Evolution python3 code

Co-evolutionary Probabilistic Structured Grammatical Evolution (Co-PSGE) is an extension to Structured Grammatical Evolution (SGE).

In Co-PSGE each individual in the population is composed by a grammar and a genotype, which is a list of dynamic lists, each corresponding to a non-terminal of the grammar containing real numbers that correspond to the probability of choosing a derivation rule. Each individual uses its own grammar to map the genotype into a program. During the evolutionary process both the grammar and the genotype are subject to variation operators.

A more in-depth explanation of the method and an analysis of its performance can be found in the article, published at the Genetic and Evolutionary Computation Conference 2022 (GECCO). If you use this code, a reference to the following work would be greatly appreciated:

@inproceedings{megane2022copsge,
author = {M\'{e}gane, Jessica and Louren\c{c}o, Nuno and Machado, Penousal},
title = {Co-Evolutionary Probabilistic Structured Grammatical Evolution},
year = {2022},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3512290.3528833},
doi = {10.1145/3512290.3528833},
}

Requirements

This code needs python3.5 or a newer version. More detail on the required libraries can be found in the requirements.txt file.

Execution

Like all grammar-based Evolutionary Algorithms, to run the algorithm to solve a problem you need a grammar and a fitness function. The folder examples/ contains the code for some benchmark problems used in Genetic Programming, and the folder grammars/ contain the respective grammars. To run, for example, a Symbolic Regression problem, you can use the following command:

python3 -m examples.symreg --grammar grammars/regression.pybnf

Parameters

The folder parameters/ contains an example of standard parameters to run. You can define the parameters on a file and specify them when executing the code. For example:

python3 -m examples.symreg --grammar grammars/regression.pybnf --parameters parameters/standard.yml

You can also add manually more parameters when calling the code without changing the parameter file. Here is an example where we define the seed:

python3 -m examples.symreg --grammar grammars/regression.pybnf --parameters parameters/standard.yml --seed 123

If you need to know the possible parameters, you can use the flag --help. For example:

python -m examples.symreg --help

Here is the list of possible parameters, and how to call them.

argument	type	description
--parameters	str	Specifies the parameters file to be used. Must include the full file extension.
--popsize	int	Specifies the population size.
--generations	int	Specifies the total number of generations.
--elitism	int	Specifies the total number of individuals that should survive in each generation.
--prob_crossover	float	Specifies the probability of crossover usage. Float required.
--prob_mutation	float	Specifies the probability of mutation usage. Float required.
--tsize	int	Specifies the tournament size for parent selection.
--min_tree_depth	int	Specifies the initialisation tree depth.
--max_tree_depth	int	Specifies the maximum tree depth.
--grammar	str	pecifies the path to the grammar file.
--grammar_probs	str	Path to file that has a list of probabilities to initialisate the grammars. Otherwise it starts with uniform distribution for each non-terminal. Json file required.
--mutate_grammar	bool	Specifies if we want the grammars to mutate.
--prob_mutation_grammar	float	Specifies the probability of occurring a mutation in the grammar of each individual.
--normal_dist_sd	float	Specifies the value of the standard deviation used in the generation of a number with a normal distribution.
--adaptive_mutation	bool	Specifies if we want to use the traditional mutation or the Adaptive Facilitated Mutation.
--prob_mutation_probs	float	Specifies the probability of occurring a mutation in the prob mutation. Option only if --adaptive_mutation is set to true.
--gauss_sd	float	Specifies the value of the standard deviation used in the generation of a number with a normal distribution. Option only if --adaptive_mutation is set to true.
--experiment_name	str	Specifies the name of the folder where stats are going to be stored.
--run	int	Specifies the run number.
--seed	float	Specifies the seed to be used by the random number generator.
--include_genotype	bool	Specifies if the genotype is to be included in the log files
--save_step	int	Specifies how often stats are saved.
--verbose	bool	Turns on the verbose output of the program.

Mutation

This code supports two types of mutations.

The standard mutation consists in changing the values in the genotype according to the PROB_MUTATION parameter.
The adaptive facilitated mutation can be enabled by setting the ADAPTIVE_MUTATION parameter to True. This mutation evolves different probabilities of mutation for each non-terminal of the grammar. Each individual contains a different array with probabilities of mutation for each non-terminal. The array starts with equal values for each non-terminal, pre-defined with the parameter PROB_MUTATION. They update each genaration based on the PROB_MUTATION_PROBS parameter, which defines the likelihood of the values suffering an alteration, and the GAUSS_SD parameter which defines the impact of those changes. The Adaptive Facilitated Mutation was publised and presented in the EuroGP 2023 conference, you can read the full paper here. If you use this mutation please cite our paper.

Support

Any questions, comments or suggestion should be directed to Jessica Mégane (jessicac@dei.uc.pt) or Nuno Lourenço (naml@dei.uc.pt).

References

O'Neill, M. and Ryan, C. "Grammatical Evolution: Evolutionary Automatic Programming in an Arbitrary Language", Kluwer Academic Publishers, 2003.

Fenton, M., McDermott, J., Fagan, D., Forstenlechner, S., Hemberg, E., and O'Neill, M. PonyGE2: Grammatical Evolution in Python. arXiv preprint, arXiv:1703.08535, 2017.

Lourenço, N., Assunção, F., Pereira, F. B., Costa, E., and Machado, P.. Structured Grammatical Evolution: A Dynamic Approach. In Handbook of Grammatical Evolution. Springer Int, 2018.

Mégane, J., Lourenço, N., and Machado, P.. Probabilistic Grammatical Evolution. In Genetic Programming, Ting Hu, Nuno Lourenço, and Eric Medvet (Eds.). Springer International Publishing, Cham, 198–213, 2021.

Carvalho, P., Mégane, J., Lourenço, N., Machado, P. (2023). Context Matters: Adaptive Mutation for Grammars. In: Pappa, G., Giacobini, M., Vasicek, Z. (eds) Genetic Programming. EuroGP 2023. Lecture Notes in Computer Science, vol 13986. Springer, Cham.

jessicamegane / co-psge