Titulo Estágio
Probabilistic Grammatical Evolution
Áreas de especialidade
Sistemas Inteligentes
Sistemas Inteligentes
Local do Estágio
ECOS-CISUC
Enquadramento
Genetic Programming (GP) is an Evolutionary Algorithm where solutions encode an algorithmic strategy, i.e., a computer program to solve a problem. These solutions can be encoded using different types of representations, such as parse-trees or grammars. The latter has gained some relevance in recent years, mainly due to Grammatical Evolution (GE). However, GE suffers from some known issues, and over the years many improvements to the traditional algorithm have been proposed in order to overcome some of these issues, namely the low locality [1,2]. Some recent works have followed a different approach, where use probabilities to encode how likely a certain symbol is to be selected to be expanded [3].
In this work, we will propose, implement and analyse new mapping mechanisms for GE. In concrete we will explore and expand on the possibility of using a genotype that encodes the probability of selecting a certain production rule.
1 - Rothlauf, F., & Oetzel, M. (2006, April). On the locality of grammatical evolution. In European Conference on Genetic Programming (pp. 320-330). Springer, Berlin, Heidelberg.
2 - Lourenço, N., Pereira, F. B., & Costa, E. (2016). Unveiling the properties of structured grammatical evolution. Genetic Programming and Evolvable Machines, 17(3), 251-289.
3 - Kim, H. T., & Ahn, C. W. (2015). A new grammatical evolution based on probabilistic context-free grammar. In Proceedings of the 18th Asia Pacific Symposium on Intelligent and Evolutionary Systems-Volume 2 (pp. 1-12). Springer, Cham.
Objetivo
The main goal of this work is to design, implement and validate a probabilistic based representation for Grammatical Evolution. The evaluation will be performed using benchmarks widely used in the literature, and we will compare the results with the state of the art variants of Grammatical Evolution.
Plano de Trabalhos - Semestre 1
1 - Review of literature
2 - Comparison between existing approaches
3 - Design and proposal of the first version of the probabilistic representation
4 - Implementation of the first version of the representation
5 - Writing of the intermediate report
Plano de Trabalhos - Semestre 2
6 - Analysis of the preliminary results.
7 - Refinement of the representation
8 - Experimental Study
9 - Scientific Article with the main results
10 - Writing of the thesis
Condições
The work is to be conducted at the ECOS laboratories of CISUC. A workplace will be provided as well as the required computational resources.
Observações
N/A
Orientador
Nuno Lourenço
naml@dei.uc.pt 📩