The accurate prediction of critical heat flux (CHF) has always been a topic of greatconcern for nuclear power plants. ln the past 30 years, many researchers have tried to usemachine learning to predict CHf, which achieved good results. This paper will review the processof predicting CHF by three machine learning methods, namely support vector machine (SVM),artificial neural network algorithm (ANN) and convolutional neural network (CNN). Theadvantages of these three methods are discussed, the opinions presented can provide referencefor this research field.