Hands-On Deep Learning for IoT

Neural network architectures

Up to now, numerous neural network architectures have been proposed and are in use. However, more or less all of them are based on a few core neural network architectures. We can categorize DL architectures into four groups:

• Deep neural networks
• Convolutional neural networks
• Recurrent neural networks
• Emergent architectures

However, DNNs, CNNs, and RNNs have many improved variants. Although most of the variants are proposed or developed for solving domain-specific research problems, the basic working principles still follow the original DNN, CNN, and RNN architectures. The following subsections will give you a brief introduction to these architectures.