MIE 2024 : IEEE WCCI2024 - CEC2024 Special Session on Make It Easy! – Evolutionary Computation with Additional Objective Functions

Organizers:
Shoichiro Tanaka (The University of Electro-Communications, Japan),
Keiki Takadama (The University of Electro-Communications, Japan),
Hiroyuki Sato (The University of Electro-Communications, Japan)

Contact email:
[email protected]

Website:
https://sites.google.com/gl.cc.uec.ac.jp/cec2024-mie

Scope and Topics:
Multi-objectivization is a new optimization paradigm that reformulates single-objective or multi-objective
optimization problems into problems with more objective functions. Adding an objective function reduces
the number of local optima and/or develops plateaus of incomparable solutions in search space, i.e.,
changes the fitness landscape. From this advantage, multi-objectivization aims to obtain more diverse and
higher-quality solutions than optimizing the original problem. However, many important issues remain
unsolved in multi-objectivization, e.g., When should problems be multi-objectivized? What and how many
objective functions should be added? Why do the additional objective functions make the problem easier?
How does the additional objective function change the landscape? To find the answer to these questions,
this special session aims to bring researchers together to explore novel methods and discuss the future
direction from the viewpoint of evolutionary computation.

The topics of this special session include but are not limited to the following topics:
- Multi-objectivization methods for continuous or combinatorial optimization problems
- Multi-objectivization methods for machine learning, dynamic or constrained optimization problems (includes studies considering other solution evaluation indicators besides the objective function value, such as novelty and diversity)
- Adaptive and dynamic multi-objectivization methods
- Empirical or theoretical analysis of the effects of additional objective functions or changing solution evaluation indicators on the algorithm performance
- Analysis of changes in the fitness landscape by the additional objective function (includes single- and multi-objective landscape analysis)
- Case studies of multi-objectivization in real-world problems