- author = "R. Muhammad Atif Azad and Conor Ryan",
- title = "Variance based selection to improve test set performance in genetic programming",
- booktitle = "GECCO '11: Proceedings of the 13th annual conference on Genetic and evolutionary computation",
- year = "2011",
- editor = "Natalio Krasnogor and Pier Luca Lanzi and Andries Engelbrecht and David Pelta and Carlos Gershenson and Giovanni Squillero and Alex Freitas and Marylyn Ritchie and Mike Preuss and Christian Gagne and Yew Soon Ong and Guenther Raidl and Marcus Gallager and Jose Lozano and Carlos Coello-Coello and Dario Landa Silva and Nikolaus Hansen and Silja Meyer-Nieberg and Jim Smith and Gus Eiben and Ester Bernado-Mansilla and Will Browne and Lee Spector and Tina Yu and Jeff Clune and Greg Hornby and Man-Leung Wong and Pierre Collet and Steve Gustafson and Jean-Paul Watson and Moshe Sipper and Simon Poulding and Gabriela Ochoa and Marc Schoenauer and Carsten Witt and Anne Auger",
- isbn13 = "978-1-4503-0557-0",
- pages = "1315--1322",
- keywords = "genetic algorithms, genetic programming",
- month = "12-16 " # jul,
- organisation = "SIGEVO",
- address = "Dublin, Ireland",
-
DOI = "
doi:10.1145/2001576.2001754",
- publisher = "ACM",
- publisher_address = "New York, NY, USA",
-
abstract = "This paper proposes to improve the performance of
Genetic Programming (GP) over unseen data by minimizing
the variance of the output values of evolving models
alongwith reducing error on the training data. Variance
is a well understood, simple and inexpensive
statistical measure; it is easy to integrate into a GP
implementation and can be computed over arbitrary input
values even when the target output is not
known.
Moreover, we propose a simple variance based selection scheme to decide between two models (individuals). The scheme is simple because, although it uses bi-objective criteria to differentiate between two competing models, it does not rely on a multi-objective optimisation algorithm. In fact, standard multi-objective algorithms can also employ this scheme to identify good trade-offs such as those located around the knee of the Pareto Front.
The results indicate that, despite some limitations, these proposals significantly improve the performance of GP over a selection of high dimensional (multi-variate) problems from the domain of symbolic regression. This improvement is manifested by superior results over test sets in three out of four problems, and by the fact that performance over the test sets does not degrade as often witnessed with standard GP; neither is this performance ever inferior to that on the training set. As with some earlier studies, these results do not find a link between expressions of small sizes and their ability to generalise to unseen data.",
- notes = "Also known as \cite{2001754} GECCO-2011 A joint meeting of the twentieth international conference on genetic algorithms (ICGA-2011) and the sixteenth annual genetic programming conference (GP-2011)",
