YPEA Toolbox for Evolutionary Algorithms in MATLAB

国际期刊International Journal of Complexity in Applied Science and Technology，收录进化计算，机器学习和大数据方面的论文， 投稿网址：https://www.inderscience.com/jhome.php?jcode=ijcast

Source Website: https://kalami.medium.com/ypea-a-toolbox-for-evolutionary-algorithms-in-matlab-6610d61a0a9c

Yarpiz Evolutionary Algorithms Toolbox

Within the last few years, I have shared many codes on Yarpiz, which implement various AI, Machine Learning, and Computational Intelligence methods, algorithms, and applications. Most of the projects are developed in MATLAB and they are also available on MATLAB File Exchange (FEX) to download.

However, there are some frequently asked questions related to the application of metaheuristics and their implementations in research projects or real-world problems. Actually, most students and researchers have difficulties defining their own problems, models, and objective functions. In many cases, we must simulate a system to calculate the key metrics and objectives for various sets of decision variables in the optimization process. This task is not straightforward and needs previous experience in implementing such projects.

To create a unified approach for implementing metaheuristics and solving optimization problems using them, I developed Yarpiz Evolutionary Algorithms Toolbox for MATLAB, or in short YPEA for MATLAB. This toolbox helps researchers and developers with these:

• It enables the user to define various types of decision variables just by describing what it is. The operations for coding and decoding variables are automatically performed by the toolbox, behind the scene.
• After the variables are defined, the user can implement the objective function, using real-world variables. This makes the implementation of objective function much easier.
• It implements several metaheuristic algorithms with similar standards, and this makes it easier to apply multiple algorithms to an optimization problem without the need to modify the code. So the problem is defined once, but it can be solved multiple times by different optimizers.