Abstract | ||
---|---|---|
This paper proposed a new adaptive control mechanism of aggregation functions (scalarizing functions) in MOEA/D, “ADaptive control of Aggregation function dePending on a search condiTion (ADAPT)”. Although MOEA/D has been well known as one of the most powerful EMO algorithms, it hasn’t been resolved which aggregation function should be choose. It is strongly depended on characteristics of the problem which aggregation function of MOEA/D is best suited and very difficult to predict which one is best suited in advance. Our proposed ADAPT changes adaptively an aggregation function of MOEA/D according to the search condition. ADAPT uses multiple aggregation functions and multiple archives corresponding to each aggregation function. The important points of ADAPT is that the number of function calls is same as that of original MOEA/D. |
Year | Venue | Field |
---|---|---|
2017 | SEAL | Multiple aggregation,Mathematical optimization,Computer science,Adaptive control |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
7 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Shinya Watanabe | 1 | 11 | 3.85 |
Takanori Sato | 2 | 0 | 1.01 |