2013 Fiscal Year Final Research Report
Development of a new multi-objective local search considering local Pareto optimality and its application to inverse problem
| Project/Area Number |
23500268
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Sensitivity informatics/Soft computing
|
|---|
| Research Institution | Muroran Institute of Technology |
|---|
Principal Investigator |
WATANABE Shinya 室蘭工業大学, 工学(系)研究科(研究院), 准教授 (30388136)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
SHIOYA Hiroyuki 室蘭工業大学, 工学研究科, 教授 (90271642)
|
|---|
| Project Period (FY) |
2011 – 2013
|
| Keywords | 進化型多目的最適化 / 局所探索 / 逆問題 / 少数投影CT / 多目的最適化アルゴリズム |
|---|
| Research Abstract |
A new multi-objective local search approach considering local Pareto optimality has been proposed. This approach also has an interpolation mechanism for capturing the whole of Pareto subsets. Through the numerical examples, the effectiveness of the proposed approach could be indicated.
Furthermore, a new approach based on Evolutionary Multi-criterion Optimization (EMO) for sparse CT problem were developed. This approach incorporates Gerchberg -Saxton algorithm (GS algorithm) that is the fact standard method in the field of phase retrieval problem as optimization tool and implements an original genetic operators utilizing the characteristics of strength distribution. The superiority of our approach could be confirmed by comparison to the existing approaches in sparse CT problem.
|
