## A Comparison of Loss Function on Deep Embedding

$Derivation: Derivatives for Common Neural Network Activation$