16.03.2021 Views

Advanced Deep Learning with Keras

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Index

A

accuracy 17

Actor-Critic (A2C) method

about 315

advantages 317

Adaptive Moments (Adam) 17

Artificial Intelligence (AI) 271

Asynchronous Advantage Actor-Critic

(A3C) 317

autoencoders

building, with Keras 74, 78-80, 83

decoder 72

encoder 72

principles 72-74

automatic colorization

autoencoders 89, 90, 95

Auxiliary Classifier GAN (ACGAN) 125,

147-150, 152-157, 161

B

backpropagation

reference 20

Batch Normalization (BN) 25, 105, 212

Bellman Equation 275

bootstrapping 288

C

Conditional GAN (CGAN) 114-205

Conditional loss function 184

Conditional VAE (CVAE) 254, 261, 263, 264

Conv2D -Batch Normalization (BN)-ReLU 51

convolution 26

Convolutional Neural Networks (CNN)

about 23, 25

convolution 26, 27

performance evaluation 28, 29

pooling operation 28

summary 28

Core Deep Learning Models

difference between 5

implementing 4, 5

critic 100

cross-domain transfer 203

CyCADA (Cycle-Consistent Adversarial Domain

Adaptation) 230

CycleGAN

about 203

principles 204, 205

using, in MNIST 227-234

using, in SVHN datasets 227-234

CycleGAN model

about 207-210

implementing, with Keras 211-227

D

decoder 72

Deep Learning (DL)

about 271

URL 97

Deep Q-network (DQN)

on Keras 296, 297

Deep Reinforcement Learning

(DRL) 271, 293-296

deep residual networks (ResNet) 49-58

denoising autoencoders (DAE) 84-86

densely connected convolutional networks

(DenseNet) 62-68

[ 347 ]

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!