Abstract
We propose and evaluate two methods for accelerating differential evolution and interactive differential evolution (IDE). The first acceleration method, which we call DE/gravity, aims to realize performance similar to that of paired-comparison-based IDE/best while removing the requirement that the IDE user must choose the best individual among all displayed individuals. The second acceleration method generates not only a conventional trial vector but also a second and third trial vector. It calculates a moving average vector, X moving, for the population between generations, and compares a given target vector with the three trial vectors of a conventional trial vector, a target vector + Xmoving, and a trial vector + Xmoving, and uses the best one among the four vectors as offspring in the next generation. We evaluate these acceleration methods and a conventional method by applying them to Gaussian mixture models and demonstrate the effectiveness of our proposed methods.
| Original language | English |
|---|---|
| Title of host publication | Proceedings - 2011 5th International Conference on Genetic and Evolutionary Computing, ICGEC 2011 |
| Pages | 287-290 |
| Number of pages | 4 |
| DOIs | |
| Publication status | Published - 2011 |
| Event | 5th International Conference on Genetic and Evolutionary Computing, ICGEC2011 - Xiamen, China Duration: Aug 29 2011 → Sep 1 2011 |
Other
| Other | 5th International Conference on Genetic and Evolutionary Computing, ICGEC2011 |
|---|---|
| Country | China |
| City | Xiamen |
| Period | 8/29/11 → 9/1/11 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Computational Theory and Mathematics
- Computer Science Applications
Cite this
Application of gravity vectors and moving vectors for the acceleration of both differential evolution and interactive differential evolution. / Funaki, Ryohei; Takagi, Hideyuki.
Proceedings - 2011 5th International Conference on Genetic and Evolutionary Computing, ICGEC 2011. 2011. p. 287-290 6042782.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Application of gravity vectors and moving vectors for the acceleration of both differential evolution and interactive differential evolution
AU - Funaki, Ryohei
AU - Takagi, Hideyuki
PY - 2011
Y1 - 2011
N2 - We propose and evaluate two methods for accelerating differential evolution and interactive differential evolution (IDE). The first acceleration method, which we call DE/gravity, aims to realize performance similar to that of paired-comparison-based IDE/best while removing the requirement that the IDE user must choose the best individual among all displayed individuals. The second acceleration method generates not only a conventional trial vector but also a second and third trial vector. It calculates a moving average vector, X moving, for the population between generations, and compares a given target vector with the three trial vectors of a conventional trial vector, a target vector + Xmoving, and a trial vector + Xmoving, and uses the best one among the four vectors as offspring in the next generation. We evaluate these acceleration methods and a conventional method by applying them to Gaussian mixture models and demonstrate the effectiveness of our proposed methods.
AB - We propose and evaluate two methods for accelerating differential evolution and interactive differential evolution (IDE). The first acceleration method, which we call DE/gravity, aims to realize performance similar to that of paired-comparison-based IDE/best while removing the requirement that the IDE user must choose the best individual among all displayed individuals. The second acceleration method generates not only a conventional trial vector but also a second and third trial vector. It calculates a moving average vector, X moving, for the population between generations, and compares a given target vector with the three trial vectors of a conventional trial vector, a target vector + Xmoving, and a trial vector + Xmoving, and uses the best one among the four vectors as offspring in the next generation. We evaluate these acceleration methods and a conventional method by applying them to Gaussian mixture models and demonstrate the effectiveness of our proposed methods.
UR - http://www.scopus.com/inward/record.url?scp=80455141514&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=80455141514&partnerID=8YFLogxK
U2 - 10.1109/ICGEC.2011.71
DO - 10.1109/ICGEC.2011.71
M3 - Conference contribution
AN - SCOPUS:80455141514
SN - 9780769544496
SP - 287
EP - 290
BT - Proceedings - 2011 5th International Conference on Genetic and Evolutionary Computing, ICGEC 2011
ER -