In recent years, the development of deep learning has opened up a new idea for researchers examining earthquake locations. Deep learning technology has been applied to earthquake locations with good results. The paper first introduces the classification of deep neural networks according to the coding and decoding of neural networks, then summarizes the basic process of deep learning. Finally, it reviews the methods of deep learning widely used in seismic locations and summarizes the characteristics and practical applications of each method. The results show that deep learning methods can help in the automatic determination of the locations of seismic events, with high accuracy of location identification, which greatly shortens the time required for the seismic location. They also have obvious advantages in processing seismic big data and can overcome some shortcomings of traditional geophysical methods in earthquake locations. It is believed that with the further development of deep learning technology, it will be more widely used in seismic location research.