Suppose we are maximizing an objective function f(x,y) via optimizing x and y.
We have to optimization algorithms:
1) Joint optimization Algorithm (JOA) that simultaneously optimizes both x and y to maximize f;
2) Separate Optimization Algorithm (SOA) that splits the process into two stages:
a) select x
b) optimize y based on the choice of x
Some SOA results give higher f than that achieved by JOA, which is relatively strange!
Does JOA always have a higher f than that of SOA?
Thank you
Dear Gamal Zayed , nice question. Alas the answer lies in the nature of your optimization problem. The notion optimize has many meanings. I like to refer to https://www.dagstuhl.de/en/program/calendar/semhp/?semnr=04461. November 7 – 12 , 2004, Dagstuhl Seminar 04461, Practical Approaches to Multi-Objective Optimization. Additionally there are several very interesting seminars dealing with your question. KR Rob
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Writing