Where Should We Stop? An Investigation on Early Stopping for GP Learning

Nguyễn, Thị Hiền; Nguyen, Xuan Phuong; McKay, Bob; Nguyen, Quang Uy

doi:10.1007/978-3-642-34859-4_39

Cited by 2 publications

(1 citation statement)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The performance of a learning system on this validation dataset may provide a useful estimate of the likely performance on out of sample data, that is, how well the learnt model may generalise beyond the training data [20]. The use of validation sets can reduce over-fitting [2], detect stasis, that is, a lack of improvement on the validation set, and can also double up as an early stopping mechanism [5] [18] in order to start a new period. Here, we propose to use a validation set to decide when to terminate the current period and also as to whether we must integrate the best individual of the current period into the joint solution.…”

Section: Validation Sets and Smart Stoppingmentioning

confidence: 99%

A New Wave

Medernach

Fitzgerald

Azad

et al. 2016

Proceedings of the Genetic and Evolutionary Computation Conference 2016

View full text Add to dashboard Cite

Wave is a novel form of semantic genetic programming which operates by optimising the residual errors of a succession of short genetic programming runs, and then producing a cumulative solution. These short genetic programming runs are called periods, and they have heterogeneous parameters. In this paper we leverage the potential of Wave's heterogeneity to simulate a dynamic evolutionary environment by incorporating self adaptive parameters together with an innovative approach to population renewal. We conduct an empirical study comparing this new approach with multiple linear regression (MLR) as well as several evolutionary computation (EC) methods including the well known geometric semantic genetic programming (GSGP) together with several other optimised Wave techniques. The results of our investigation show that the dynamic Wave algorithm delivers consistently equal or better performance than Standard GP (both with or without linear scaling), achieves testing fitness equal or better than multiple linear regression, and performs significantly better than GSGP on five of the six problems studied.

show abstract

Section: Validation Sets and Smart Stoppingmentioning

confidence: 99%