Research on Athlete Performance Prediction Model Based on Recurrent Neural Network

Recurrent neural network is an important model in the field of deep learning. Similar network structure is used to recursively form a more complex deep network with a relatively simple structure. By adding extra weights to the network to create cycles in the network graph, and using long- distance dependence information, high prediction accuracy can be achieved with sufficient data. The training speed of recurrent neural networks cannot be improved and the gradient gradually disappears. LSTM improves the hidden layer nodes of RNN into special cell structures, which can perform better in longer sequences. In this paper, the cyclic neural network algorithm and its improved model based on LSTM are studied, and the athletes' performance prediction model based on cyclic neural network is constructed, which can be used to predict athletes' performance, and high prediction accuracy can be achieved when the amount of data is sufficient.