This chapter surveys the current research directions of pattern recognition in sports by resorting to multiple classification methods, including non-sequence classification, such as support vector machines and neural network approaches. Also sequence classification as obtained by ensemble learning and recurrent neural networks are discussed. It analyses the use of these artificial intelligence methods to autonomously recognize actions and performance signatures, at both individual and collective levels, presenting their advantages and drawbacks, performance metrics required as representative features, and usage examples. It also discusses the research challenges, such as the individual profile of the athletes, especially at the physiological level, and technical limitations, as well as the remaining open research questions. This chapter considers multiple examples, from individual to collective sports.