How do you know if your Autoencoder network is fully connected?

I am new in deep learning and I am confused about fully connected network. Is an Autoencoder with more than one hidden layers a type of deep neural network (DNN)? Is DNN always fully connected? Let's say an Autoencoder network with three hidden layers was built using H2O R package, how do you know if the Autoencoder network that you build is fully connected?

Is Autoencoder with more than one hidden layers a type of deep neural network (DNN)?

Yes.

Is DNN always fully connected?

No. For example Echo State Networks are sparsely connected DNNs.

how do you know if the Autoencoder network that you build is fully connected?

I am not sure about R because I don't use it but for Keras/Python you can read the layer descriptions. The most common predefined layers (Dense, Convolutional, RNN, LSTM..) are fully connected. If you want to have a sparsely connected layer, then you can probably create a custom one using a Lambda Layer.

• Some people refer to fully-connected layers meaning Dense layers, which would not include convolutional or recurrent layers. – n1k31t4 Aug 7 '18 at 9:48