A weather-balloon photo hits an input layer of pixel-like values, passes through one or more hidden layers of nodes, and leaves an output layer as an “icing-risk” score. In a fully connected net, each node talks to every node in the next layer, and layer widths may differ. What structure should the tester name?

Short Explanation

Pixel-like values into an input layer, through one or more hidden layers, out as an icing-risk score: that is input, hidden, and output. Fully connected nets of that form are multi-layer perceptrons. Hidden layers are part of a DNN.

Full Explanation

DNN structure is input, hidden, and output layers with typical full connectivity. Dropping hidden layers, collapsing to a one-layer perceptron, or turning the item into an implementation task misses that structure.

×