Learning, evolution and generalisation

Kushchu, Ibrahim

doi:10.1109/cec.2003.1299394

Cited by 4 publications

(2 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A model is overfit if it is specifically tuned to training instances, which means the model will be accurate at predicting the values of those instances, but poor at predicting values of unseen instances. Conventional machine learning uses learning as a paradigm, and evolutionary computation uses evolution as a paradigm; learning is adaptation at the individual level, and evolution is adaptation at the population level [52]. To prevent overfitting, one usually just splits the instance space between training and testing, to validate training models.…”

Section: Chapter 2 Background and Related Workmentioning

confidence: 99%

Modelling Evolvability in Genetic Programming

Fowler

Banzhaf

2016

Lecture Notes in Computer Science

View full text Add to dashboard Cite

We develop a tree-based genetic programming system, capable of modelling evolvability during evolution through artificial neural networks (ANN) and exploiting those networks to increase the generational fitness of the system. This thesis is empirically focused; we study the effects of evolvability selection under varying conditions to demonstrate the effectiveness of evolvability selection. Evolvability is the capacity of an individual to improve its future fitness. In genetic programming (GP), we typically measure how well a program performs a given task at its current capacity only. We improve upon GP by directly selecting for evolvability. We construct a system, Sample-Evolvability Genetic Programming (SEGP), that estimates the true I thank ACENET for material support. Experiments were conducted on hardware provided by ACENET, in partnership with Compute Canada, a high performance computing (HPC) consortium for universities based in the Atlantic Provinces of Canada. ACENET is funded by the Atlantic Provinces Opportunities Agency, the

show abstract

Section: Chapter 2 Background and Related Workmentioning

confidence: 99%

Modelling Evolvability in Genetic Programming

Fowler

Banzhaf

2016

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…While learning takes place based on this prediction error, the generality of the resulting function is important. For this reason, it is necessary not only to carry out learning using training data, but also to evaluate the resulting function with the test data [6]. The number of training data sets provided was 1,210.…”

Section: Task Settingmentioning

confidence: 99%

Evolutionary construction of a simulator for real robots

Kamio

Iba

Proceedings of the 2004 Congress on Evolutionary Computation (IEEE Cat. No.04TH8753)

View full text Add to dashboard Cite

Abstract-In order to acquire useful motions of a real-world robot, it is necessary to carry out learning in a real environment. However, learning is difficult within a real environment. In addition, the acceleration of learning is required for a practical execution.In this paper, we propose an approach to the learning acceleration using data retrieved from the real environment. This consists of the method of automatically constructing the simulator from real data and of learning a robot controller with the simulator. The experimental results suggest that our GPbased technique enables the effective controller learning.

show abstract

Third-order generalization: A new approach to categorizing higher-order generalization

Neville

2008

Neurocomputing

View full text Add to dashboard Cite

Learning, evolution and generalisation

Cited by 4 publications

References 24 publications

Modelling Evolvability in Genetic Programming

Modelling Evolvability in Genetic Programming

Evolutionary construction of a simulator for real robots

Third-order generalization: A new approach to categorizing higher-order generalization

Contact Info

Product

Resources

About