Pro Deep Learning with TensorFlow: A Mathematical Approach to Advanced Artificial Intelligence in Python

Index

A

1. Activation functions, neuron/perceptron

   1. binary threshold activation function

   2. linear activation function

   3. rectified linear unit

   4. sigmoid activation function

   5. SoftMax activation function

   6. tanh activation function

2. AdadeltaOptimizer

3. AdagradOptimizer

4. AdamOptimizer

5. Auto encoders

   1. architecture

   2. cases

   3. combined classification network, class prediction

   4. denoising auto-encoder implementation

   5. element wise activation function

   6. hidden layer

   7. KL divergence

   8. learning rule of model

   9. multiple hidden layers

   10. network, class prediction

   11. sparse

   12. unsupervised ANN

B

1. Backpropagation

   1. convolution layer

   2. for gradient computation

      1. cost derivative

      2. cost function

      3. cross-entropy cost, SoftMax activation layer

      4. forward pass and backward pass

      5. hidden layer unit

      6. independent sigmoid output units

      7. multi-layer neural network