Wrap-up

Advanced Deep Learning with Keras

Zach Deane-Mayer

Data Scientist

So far...

Functional API
Shared layers
Categorical embeddings
Multiple inputs
Multiple outputs
Regression / Classification in one model

Shared layers

Useful for making comparisons

Basketball teams
Image similarity / retrieval
Document similarity

Known in the academic literature as Siamese networks

Link to blog post
Link to academic paper

Multiple inputs

Multiple outputs

Skip connections

input_tensor = Input((100,))
hidden_tensor = Dense(256, activation='relu')(input_tensor)
hidden_tensor = Dense(256, activation='relu')(hidden_tensor)
hidden_tensor = Dense(256, activation='relu')(hidden_tensor)
output_tensor = Concatenate()([input_tensor, hidden_tensor])
output_tensor = Dense(256, activation='relu')(output_tensor)

Visualizing the Loss Landscape of Neural Nets

Best of luck!

Advanced Deep Learning with Keras