Each evolutionary algorithm received the ideas from different biology environment. We simplified the complex biology evolutionary process into evolutionary algorithms. For investigating the reason why those algorithms can reach the goal, we turn to mathematical weapons for help. Sometimes, the mathematical made me very confused. Once eliminate the concrete definitions in each type of algorithms. The algorithms may seem to be very similar. All algorithms get the same idea – try to create a relativity simple system from a very complex system, then we find result from these small system. The problem is - why all bio-inspired algorithms get this same idea? Is this a natural law? Just like nature biology creator tries to separate the environment from inner to the world outside? It is a funny coincide.

In fact, we can use this coincide to create some more robust & efficient algorithms. This coincide illustrates the interconnected between evolutionary algorithms. Image that, there is a set contains all kinds of evolutionary algorithms. We can define some relationship between each algorithm. In mathematical language, we call it a ‘space’. This is a space created by evolutionary algorithm! In this space constructed by evolutionary algorithms, we do not need to find novel evolutionary algorithms one by one. We do not need to design evolutionary algorithms for each real world scenario. We only need to know how to approximating a kind of algorithm we want! In this way, we even can find some algorithms can’t called ‘bio-inspired’, but meet our demanding!

However, there are many very hard problems. First, the ‘space’ isn’t a normal one. Can we apply traditional mathematical idea to create this ‘space’? Moreover, evolutionary algorithms always contained many type of mathematical ideas such as ‘Markov Process’, ‘Group theory’, ‘Differential Operators’. To organize all algorithms into universal one seems an impossible mission. How can we find an alternative way to create this ‘evolutionary algorithms space’? Finally, natural always very hard to forecast, we must face chaos, uncertain & complexity. The measure is very difficulty, how can we measure it.

But I think there are still many standpoints focus us to investigate it. Nowadays, we always get intelligent from nature. But we more like create natural. A natural in computer only created from human. It songs good, isn’t it? So, this is the reason I like evolutionary algorithms most. I not only become a scientist for digging up the hidden law in the beautiful world, but also an engineer for solving real problems human faced