Python: Advanced Guide to Artificial Intelligence

The DenseNet model

DenseNet is an extension of ResNet proposed by Huang et al. (https://arxiv.org/pdf/1608.06993.pdf). In ResNet blocks, the previous layer is merged into the future layer by summation. In DenseNet, the previous layer is merged into the future layer by concatenation. DenseNet connects all the layers to the previous layers and the current layer to the following layers.

In the following diagram, it can be seen how the feature maps are supplied as input to the other layers: 

Reproduced with permission from Huang et al.

This way, it provides several advantages such as smoother gradients, feature transformation and so on. This also ...