Untapped Potential of Genetic Programming: Transfer Learning and Outlier Removal
In the era of Deep Learning and Big Data, the place of Genetic Programming (GP) within the Machine Learning area seems difficult to define. Whether it is due to technical constraints or conceptual barriers, GP is currently not a paradigm of choice for the development of state-of-the-art machine learning systems. Nonetheless, there are important features of the GP approach that make it unique and should continue to be actively explored and studied. In this work we focus on two aspects of GP that have previously received little or no attention, particularly in tree-based GP for symbolic regression. First, on the potential of GP to perform transfer learning, where solutions evolved for one problem are transferred to another. Second, on the potential of GP individuals to detect the true underlying structure of an input dataset and detect anomalies in the input data, what are known as outliers. This work presents initial results on both issues, with the goal of fostering discussion and showing that there is still untapped potential in the GP paradigm.
This research was funded by CONACYT (Mexico) Fronteras de la Ciencia 2015-2 Project No. FC-2015-2/944 and TecNM project no. 6826-18-P, and first and third authors were respectively supported by CONACYT graduate scholarship No. 302526 and No. 573397.
- 3.Chitty, D.M.: Faster GPU based genetic programming using A two dimensional stack. CoRR abs/1601.00221 (2016)Google Scholar
- 6.Floreano, D., Mattiussi, C.: Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies. MIT Press (2008)Google Scholar
- 9.Galván-López, E., Vazquez-Mendoza, L., Schoenauer, M., Trujillo, L.: On the Use of Dynamic GP Fitness Cases in Static and Dynamic Optimisation Problems. In: EA 2017- International Conference on Artificial Evolution, pp. 1–14. Paris, France (2017)Google Scholar
- 11.Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning. MIT Press (2016)Google Scholar
- 13.Kotanchek, M., et al.: Pursuing the Pareto Paradigm: Tournaments, Algorithm Variations and Ordinal Optimization, pp. 167–185. Springer US (2007)Google Scholar
- 14.López, U., Trujillo, L., Martinez, Y., Legrand, P., Naredo, E., Silva, S.: RANSAC-GP: Dealing with Outliers in Symbolic Regression with Genetic Programming, pp. 114–130. Springer International Publishing, Cham (2017)Google Scholar
- 16.McConaghy, T.: Genetic Programming Theory and Practice IX, chap. FFX: Fast, Scalable, Deterministic Symbolic Regression Technology, pp. 235–260. Springer New York, New York, NY (2011)Google Scholar
- 17.Miranda, L.F., Oliveira, L.O.V.B., Martins, J.F.B.S., Pappa, G.L.: How noisy data affects geometric semantic genetic programming. In: Proceedings of the Genetic and Evolutionary Computation Conference, GECCO ’17, pp. 985–992. ACM, New York, NY, USA (2017)Google Scholar
- 18.Moraglio, A., Krawiec, K., Johnson, C.G.: Parallel Problem Solving from Nature - PPSN XII: 12th International Conference, Taormina, Italy, September 1–5, 2012, Proceedings, Part I, chap. Geometric Semantic Genetic Programming, pp. 21–31. Springer Berlin Heidelberg, Berlin, Heidelberg (2012)Google Scholar
- 19.Muñoz, L., Silva, S., Trujillo, L.: M3GP: multiclass classification with GP. In: P. Machado, et al. (eds.) 18th European Conference on Genetic Programming, LNCS, vol. 9025, pp. 78–91. Springer, Copenhagen (2015)Google Scholar
- 20.Muñoz, L., Trujillo, L., Silva, S., Vanneschi, L.: Evolving multidimensional transformations for symbolic regression with m3gp. Memetic Computing (2018). https://doi.org/10.1007/s12293-018-0274-5
- 25.Spector, L.: Assessment of problem modality by differential performance of lexicase selection in genetic programming: a preliminary report. In: Proceedings of the fourteenth international conference on Genetic and evolutionary computation conference companion, GECCO Companion ’12, pp. 401–408. ACM (2012)Google Scholar
- 26.Tran, C.T., Zhang, M., Andreae, P., Xue, B.: Genetic programming based feature construction for classification with incomplete data. In: Proceedings of the Genetic and Evolutionary Computation Conference, GECCO ’17, pp. 1033–1040. ACM, New York, NY, USA (2017)Google Scholar