IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION

Please respect the original author of this text and indicate the source! Can the entire text be copied as is?

"This thing called evolutionary algorithm, it's true that it's effective, and it's also true that it's not effective at all.
Here's an explanation to elevate its status.
1. It is a general algorithm that can solve any problem.
2. For problems with low efficiency in solving, if you solely rely on evolutionary algorithms, the efficiency is indeed low compared to traditional algorithms and mathematical algorithms. This is because traditional algorithms and mathematical methods are specifically designed for specific problems. But for evolutionary algorithms, you haven't done anything! You can also add artificially designed heuristic factors within this algorithm, and the efficiency of the final hybrid algorithm will only be higher, not lower.
3. The idea behind this thing is actually consistent with reinforcement learning. With the emergence of AlphaGo, who dares to say that reinforcement learning is useless? How does reinforcement learning work? Random sampling and using the sampled data to update the strategy, and then using the strategy to obtain new samples. How does evolutionary algorithm work? Randomly generating initial solutions, and updating and generating new solutions based on existing solutions. Do you think these two are fundamentally different?
This thing is awesome, but due to its low entry barrier, it is mixed with too many inferior products, which lowers the overall standard."

I wrote an article this year, and after my advisor read it, he thought I could give TEVC a try. The journal's speed is quite fast, as the editor sent it for external review in less than a week. After three months of external review, it was returned with a rejection.
By chance, I found out who one of the reviewers was, and indeed, they are a top expert in the industry. Through comparing their reviewer comments with their own published articles, I discovered some issues. I want to share them to prevent my colleagues from making the same mistakes:
1. It seems beneficial to have connections with influential experts. I read a top journal article from an ordinary university student in China, and it didn't impress me much. However, because it had this influential expert as a co-author, it seemed to have some impact.
2. It appears that journals like TEVC may not solely focus on the quality of the article, as some colleagues have mentioned. The reviewer raised some points that were present in their own article. For example, I used the same comparison algorithm data as in one of their recent articles, yet they claimed my comparison algorithms were outdated. It seems a bit hypocritical.
3. It is important to strictly follow the experimental analysis process of the journal. Although it doesn't seem explicitly stated anywhere (at least I haven't found it), it is advisable to reference published articles to avoid the situation I faced, where the reviewer used it against me.