進化計算, ソフトコンピューティング, 人工知能, 最適化
Evolutionary computation (EC) algorithms, as a branch of optimization technology, borrow ideas from biological evolution and simulate the survival of the fittest repeatedly to find the global optimum.
They have attracted the attention of many researchers and have successfully solved lots of real-world applications thanks to their numerous advantages.
The demand for EC algorithms with high performance grows rapidly along with real-world problems that need to be solved has become rather complicated.
Thus, the primary focus of our lab is to find ways of accelerating EC convergence with lower cost consumption from three research directions below.
- Accelerating EC/interactive EC with estimated convergence point(s);
- Proposing new search strategies to improve the performance of existing EC algorithms;
- Developing new powerful EC algorithms.
Not limited to the topics mentioned above, we are also actively exploring some research directions. Here, we list several interesting topics that will be covered.
- Developing EC algorithms for practical problems with strong constraints and high computational costs;
- Analyzing fitness landscape characteristics and extract hidden information;
- Fusion of multiple optimization technologies, such as EC, fuzzy logic system, and artificial neural networks;
- Fusion of human sensibility and computing power;
- Reducing user fatigue in interactive EC;
We always welcome collaboration with enthusiastic researchers from various fields.