Volume II – Deep Learning
1. Introduction to Artificial Neural Networks with Tensorflow and Keras
1. From biological to artificial neurons: Biological neurons, logical
computations with neurons, the perceptron, the multilayer perceptron
and backpropagation, MLPs for regression, MLPs for classification.
2. Implementing MLPs with Tensorflow and Keras: Installing Tensorflow
and Keras, building an image classifier using the Sequential API,
building a regression MLP using the Sequential API, building complex Volume II – Deep Learning
1. Introduction to Artificial Neural Networks with Tensorflow and Keras
1. From biological to artificial neurons: Biological neurons, logical
computations with neurons, the perceptron, the multilayer perceptron
and backpropagation, MLPs for regression, MLPs for classification.
2. Implementing MLPs with Tensorflow and Keras: Installing Tensorflow
and Keras, building an image classifier using the Sequential API,
building a regression MLP using the Sequential API, building complex 8. Classification and Localization
9. Object Detection: Fully convolutional networks, You Only Look Once
(YOLO)
10. Semantic Segmentation
6. Processing Sequences Using RNNs and LSTMs
1. Recurrent Neurons and Layers: Memory cells, input and output
sequences
2. Training RNNs
3. Forecasting a Time Series: baseline metrics, implementing a simple
RNNs, deep RNNs, forecasting several time steps ahead
4. Handling Long Sequences: fighting the unstable gradients problem,
tackling the short-term memory problem.
7. Natural Language Processing with RNNs
1. Generating Text Using a Character RNN: creating the training dataset,
how to split a sequential dataset, chopping the sequential dataset into
multiple windows, building and training the char-RNN model, using the
char-RNN model, generating fake text, stateful RNN.
2. Sentiment Analysis: masking, reusing pre-trained embedding
3. An Encoder-Decoder Network for Neural Machine Translation:
bidirectional RNNs, beam search.
4. Attention Mechanisms: visual attention, the transformer architecture
for attention.
5. Recent Innovations in Language Models.
8. Representation Learning and Generative Learning Using Autoencoders and
GANs
1. Efficient Data Representation
2. Performing PCA with an Undercomplete Linear Autoencoder
3. Stacked Autoencoder: implementing a stacked autoencoder using
Keras, visualizing the reconstructions, visualizing the fashion MNIST
dataset, unsupervised pretraining using attacked autoencoders, tying
weights, training one autoencoder at a time.
4. Convolutional Autoencoders
5. Recurrent Autoencoders
6. Denoising Autoencoders
7. Sparse Autoencoders
8. Variational Autoencoders: generating fashion MNIST images
9. Generative Adversarial Networks: the difficulties of training GANS,
deep convolutional GANs, progressive growing of GANs, StyleGANs.
9. Reinforcement Learning
1. Learning to Optimize Rewards
2. Policy Search
3. Introduction to OpenAI Gym
4. Neural Network Policies
5. Evaluating Actions: The Credit Assignment Problem
6. Policy Gradients
7. Markov Decision Processes
8. Temporal Difference Learning
9. Q-Learning: exploration policies, approximate Q-learning and deep Qlearning
10. Implementing Deep Q-Learning 11. Deep Q-Learning Variants: fixed Q-value targets, double DQN,
prioritized experience replay, dueling DQN
12. The TF-Agents Library: installing TF-agents, TF-agents environments,
environment specifications, environment wrappers and Atari
preprocessing, training architecture, creating the deep Q-network,
creating the DQN agent, creating the replay buffer and the
corresponding observer, creating training metrics, creating training
metrics, creating the collect driver, creating the dataset, creating the
training loop.
13. Overview of Some Popular RL Algorithms
10. Training and Deploying Tensorflow Models at Scale
1. Serving a Tensorflow Model: using a Tensorflow model, creating a
prediction service on the GCP AI platform, using the prediction service
2. Deploying Model to a Mobile or Embedded Device
3. Using GPUs to Speed Up Computations: getting the GPU, using a GPUequipped virtual machine, Google Collaboratory, managing GPU RAM,
placing operations and variables on devices, parallel execution across
multiple devices.
4. Training Models across Multiple Devices: model parallelism, data
parallelism, training at scale using distribution strategies API, training a
model on a Tensorflow cluster, running large training jobs on Google
cloud AI platform, black-box hyperparameter tuning AI platform