TY - GEN
T1 - ESAGP – A semantic GP framework based on alignment in the error space
AU - Ruberto, Stefano
AU - Vanneschi, Leonardo
AU - Castelli, Mauro
AU - Silva, Sara
N1 - Ruberto, S., Vanneschi, L., Castelli, M., & Silva, S. (2014). ESAGP – A semantic GP framework based on alignment in the error space. In P. García-Sánchez, J. J. Merelo, V. M. Rivas Santos, M. Nicolau, K. Krawiec, M. I. Heywood, M. Castelli, ... K. Sim (Eds.), Genetic Programming - 17th European Conference, EuroGP 2014, Revised Selected Papers (pp. 150-161). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8599). Springer Verlag. https://doi.org/10.1007/978-3-662-44303-3_13
PY - 2014/1/1
Y1 - 2014/1/1
N2 - This paper introduces the concepts of error vector and error space, directly bound to semantics, one of the hottest topics in genetic programming. Based on these concepts, we introduce the notions of optimally aligned individuals and optimally coplanar individuals. We show that, given optimally aligned, or optimally coplanar, individuals, it is possible to construct a globally optimal solution analytically. Thus, we introduce a genetic programming framework for symbolic regression called Error Space Alignment GP (ESAGP) and two of its instances: ESAGP-1, whose objective is to find optimally aligned individuals, and ESAGP-2, whose objective is to find optimally coplanar individuals. We also discuss how to generalize the approach to any number of dimensions. Using two complex real-life applications, we provide experimental evidence that ESAGP-2 outperforms ESAGP-1, which in turn outperforms both standard GP and geometric semantic GP. This suggests that “adding dimensions” is beneficial and encourages us to pursue the study in many different directions, that we summarize in the final part of the manuscript.
AB - This paper introduces the concepts of error vector and error space, directly bound to semantics, one of the hottest topics in genetic programming. Based on these concepts, we introduce the notions of optimally aligned individuals and optimally coplanar individuals. We show that, given optimally aligned, or optimally coplanar, individuals, it is possible to construct a globally optimal solution analytically. Thus, we introduce a genetic programming framework for symbolic regression called Error Space Alignment GP (ESAGP) and two of its instances: ESAGP-1, whose objective is to find optimally aligned individuals, and ESAGP-2, whose objective is to find optimally coplanar individuals. We also discuss how to generalize the approach to any number of dimensions. Using two complex real-life applications, we provide experimental evidence that ESAGP-2 outperforms ESAGP-1, which in turn outperforms both standard GP and geometric semantic GP. This suggests that “adding dimensions” is beneficial and encourages us to pursue the study in many different directions, that we summarize in the final part of the manuscript.
UR - http://www.scopus.com/inward/record.url?scp=84927623396&partnerID=8YFLogxK
U2 - 10.1007/978-3-662-44303-3_13
DO - 10.1007/978-3-662-44303-3_13
M3 - Conference contribution
AN - SCOPUS:84927623396
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 150
EP - 161
BT - Genetic Programming - 17th European Conference, EuroGP 2014, Revised Selected Papers
A2 - García-Sánchez, Pablo
A2 - Merelo, Juan J.
A2 - Rivas Santos, Victor M.
A2 - Nicolau, Miguel
A2 - Krawiec, Krzysztof
A2 - Heywood, Malcolm I.
A2 - Castelli, Mauro
A2 - Sim, Kevin
PB - Springer Verlag
T2 - 17th European Conference on Genetic Programming, EuroGP 2014
Y2 - 23 April 2014 through 25 April 2014
ER -