Loading...
Please wait, while we are loading the content...
Similar Documents
Sizing the Population with Respect to the Local Progress in (1, lambda)-Evolution Strategies - A Theoretical Analysis (1995)
| Content Provider | CiteSeerX |
|---|---|
| Author | Hansen, Nikolaus Gawelczyk, Andreas Ostermeier, Andreas |
| Description | In Proceedings of the 1995 IEEE Conference on Evolutionary Computation This paper presents an analysis of the local serial rate of progress with respect to the number of offspring for the (1,)-evolution strategy. It is shown that local serial progress is maximized when the expected progress of the second best offspring is zero. The theoretical results lead to a simple but efficient adaptation rule for , which needs no extra fitness function evaluations and only small computational expense. Simulations of the -adaptation on simple test functions are shown. Keywords: evolutionary algorithms, evolution strategy; population size, offspring number; adaptation, strategy parameters; progress, serial progress In 1995 IEEE International Conference on Evolutionary Computation Proceedings: 80--85. 1. Introduction In most applications of evolution strategies (ESs), the number of offspring is held fixed throughout optimization. Based on results yielded for a spherical objective function in [7], its value can be chosen such that the serial rate of progress (... |
| File Format | |
| Language | English |
| Publisher | IEEE Press |
| Publisher Date | 1995-01-01 |
| Access Restriction | Open |
| Subject Keyword | Strategy Parameter Serial Progress Local Serial Rate Serial Rate Efficient Adaptation Rule Theoretical Result Population Size Theoretical Analysis Ieee International Conference Evolution Strategy Expected Progress Evolutionary Algorithm Simple Test Function Small Computational Expense Local Progress Local Serial Progress Evolutionary Computation Proceeding Extra Fitness Function Evaluation Spherical Objective Function |
| Content Type | Text |
| Resource Type | Article |
