ABC of Deep Learning (Part 4 of 5)

Network Architecture refers to the overall neural network structure, including the number of layers, the number of neurons in each layer, and the connections between them. There is not a single best network architecture for all problems; the best architecture depends on the specific problem being solved. This is also a very active area of research, with new architectures being constantly proposed. Some wildly successful architectures have been proposed in recent years, such as convolutional neural networks and recurrent neural networks. The most advanced architectures are usually based on a combination of these basic types.