Pattern recognition needs very good probability theory and mathematical statistics; In addition, a small amount of matrix algebra, random processes and some high-number operations will be used, which is of course relatively basic; I'm afraid you need to learn functional if you want to go deeper, but generally you don't need to reach this depth. Intelligent algorithms such as neural network and genetic algorithm have very important applications in pattern recognition, but generally it is not necessary to learn artificial intelligence in computer science. We control an interdisciplinary subject called intelligent control, and that's all. Intelligent control does not need any foundation, as long as there is a collection of middle school mathematics and a little space understanding. The basis of fuzzy mathematics is included in this subject.