Abstract: The hot deformation behavior of silicon carbide particles reinforced aluminum matrix composite (SiCp/Al) was investigated by isothermal compression testing at temperatures of 350 ℃, 400 ℃, 450 ℃, and 450 ℃ as well as strain rates of 0.001~1 s^-1 on a Gleeble-3180 simulator. Based on the true strain-true stress data of hot compression experiment, the modified Johnson-Cook constitutive model (M-JC) was constructed, considering the coupling effects of strain, strain rate and deformation temperature on flow stress, and the Artificial neuron network models (ANN) was established. The results show that the flow stress of SiCp/2014Al composite increases with the increase of stain rate and the decrease of temperature. Compared with the M-JC model, the ANN model has a lower root mean square error of 0.51 MPa, a lower average absolute error of 1.43%, and a higher correlation coefficient (0.999 7), which indicates that the ANN that it has higher prediction accuracy and reliability in the prediction of the hot deformation behavior of SiCp/2014Al composite.