:::

演講公告
  • 標題:Control Parameter Settings in Differential Evolution for Global Optimization Problems
  • 公告日期:2017-04-24

 

Speaker: 林豐澤教授(中國文化大學應數系)

Title: Control Parameter Settings in Differential Evolution for Global Optimization Problems
 
Time: 16:10–17:00, Monday, May 1, 2017
 
Venue: Room 070221, 2F, Zhixi Building (志希樓2 樓e 化教室)
 
Refreshment: Refreshments will be served in Faculty Lounge 30 minutes before the talk.

Abstract: Differential evolution (DE) is one of the most powerful
stochastic real-parameter optimization methods, which offers
a way of optimizing a problem without using its gradient.
Over the past years, DE has been successfully applied to
resolve combinatorial optimization problems, which exhibits
remarkable performance in optimizing a wide variety of multidimensional
and multi-modal objective functions. However,
the DE performance primarily depends on the mutation
strategy, the crossover operator and the control parameters.
In this talk, I would like to focus on discussing how to set
the control parameter values in DE for global optimization
problems. The control parameters include population size
NP, scaling factor F, and crossover probability CR. I will
present several experimental results on how control parameters
are being changed during the evolutionary process on global
optimization problems. Then, I will propose a parameter
control method based on the adaptation of three control
parameters, associated with the evolutionary process on global
optimization.

  • 附件檔案: 演講公告 Adobe PDF
  • 參考連結:
  • 張貼人:guavawei
  • 最後修改時間:2017-04-24 PM 4:44

cron web_use_log