CodePudding user response:
Neural network has many variants, such as within DNN: deep neural network, the depth of the neural network; CNN: convolutional neural network, the convolutional neural network, mainly used in the image, computer vision field, such as image recognition, target detection, video detection, etc. RNN: mainly used for processing sequence data, such as voice: speech recognition, speech separation; Natural language processing field (text detection, machine translation), used for the field of video image; GAN: emergent against network, also belong to a kind of neural network, solve the problem of a wide range of aspects, not limited to voice, image, video, text, enhance learning (learning), such as Alpha - go successful defeat the strongest master excel, Alpha - zero short-term self-study go beat it brother Alpha - go, so, in summary, the neural network to solve the problem of aspect mainly include: - computer vision, including image and video) - field of voice (voice recognition, etc.) - natural language processing (mainly take the text) - intelligent driving, wisdom city, medical - the other side of life... Can say that life in the way of communication is written communication (E-mail, letter), voice communication (on the phone, instant voice) and use images, video and other forms as a result, the neural network (exact deep learning) can solve the problem of actually covers all aspects of life,