Understanding Deep Neural Networks Training Course

Part 1 – Deep Learning and DNN Concepts

Introduction to AI, Machine Learning & Deep Learning

• The history, core concepts, and typical applications of artificial intelligence, distinct from sci-fi fantasies.
• Collective Intelligence: The aggregation of knowledge shared by numerous virtual agents.
• Genetic algorithms: Evolving a population of virtual agents through selection processes.
• Machine Learning: Fundamental definitions.
• Task types: Supervised learning, unsupervised learning, and reinforcement learning.
• Action types: Classification, regression, clustering, density estimation, and dimensionality reduction.
• Examples of Machine Learning algorithms: Linear regression, Naive Bayes, and Random Trees.
• Machine Learning vs. Deep Learning: Identifying problems where Machine Learning (e.g., Random Forests & XGBoosts) remains the state of the art.

Basic Concepts of a Neural Network (Application: Multi-layer Perceptron)

• A review of the mathematical foundations.
• Defining a neural network: Classical architecture, activation functions,
• Weighting of previous activations and network depth.
• Defining network learning: Cost functions, back-propagation, Stochastic Gradient Descent, and maximum likelihood.
• Modeling neural networks: Structuring input and output data based on the problem type (regression, classification, etc.). Understanding the curse of dimensionality.
• Distinguishing between multi-feature data and signals. Selecting the appropriate cost function based on data characteristics.
• Function approximation by neural networks: Theory and practical examples.
• Distribution approximation by neural networks: Theory and practical examples.
• Data Augmentation: Techniques for balancing datasets.
• Generalizing the results of neural networks.
• Network initialization and regularization: L1 / L2 regularization and Batch Normalization.
• Optimization and convergence algorithms.

Standard ML / DL Tools

This section provides an overview of key tools, highlighting their advantages, disadvantages, ecosystem positioning, and usage.

• Data management tools: Apache Spark and Apache Hadoop.
• Machine Learning: Numpy, Scipy, and Sci-kit.
• High-level DL frameworks: PyTorch, Keras, and Lasagne.
• Low-level DL frameworks: Theano, Torch, Caffe, and TensorFlow.

Convolutional Neural Networks (CNN)

• Overview of CNNs: Fundamental principles and applications.
• Basic CNN operations: Convolutional layers and the use of kernels,
• Padding & stride, feature map generation, and pooling layers. Extensions to 1D, 2D, and 3D.
• Review of CNN architectures that have established the state of the art in classification.
• Image processing: LeNet, VGG Networks, Network in Network, Inception, and ResNet. Analysis of innovations introduced by each architecture and their broader applications (e.g., 1x1 Convolution or residual connections).
• Utilization of attention models.
• Application to common classification tasks (text or image).
• CNNs for generation: Super-resolution and pixel-to-pixel segmentation.
• Primary strategies for enhancing feature maps for image generation.

Recurrent Neural Networks (RNN)

• Overview of RNNs: Fundamental principles and applications.
• Basic RNN operations: Hidden activations, back-propagation through time, and the unfolded version.
• Evolution towards Gated Recurrent Units (GRUs) and LSTM (Long Short-Term Memory).
• Review of different states and the advancements brought by these architectures.
• Challenges related to convergence and vanishing gradients.
• Classical architectures: Time series prediction and classification.
• RNN Encoder-Decoder architectures. Use of attention models.
• NLP applications: Word / character encoding and translation.
• Video applications: Predicting the next frame in a video sequence.

Generative Models: Variational Auto-Encoder (VAE) and Generative Adversarial Networks (GAN)

• Introduction to generative models and their relationship with CNNs.
• Auto-encoders: Dimensionality reduction and limited generation capabilities.
• Variational Auto-encoders: Generative modeling and distribution approximation. Definition and use of latent space. The reparameterization trick. Applications and observed limitations.
• Generative Adversarial Networks: Fundamental concepts.
• Dual Network Architecture (Generator and Discriminator) with alternating learning and available cost functions.
• GAN convergence and associated difficulties.
• Improved convergence techniques: Wasserstein GAN, Began, and Earth Mover's Distance.
• Applications in image/photograph generation, text generation, and super-resolution.

Deep Reinforcement Learning

• Introduction to reinforcement learning: Controlling an agent within a defined environment
• Characterized by states and possible actions.
• Using neural networks to approximate the state function.
• Deep Q Learning: Experience replay and its application to video game control.
• Policy optimization. On-policy & off-policy methods. Actor-Critic architecture. A3C.
• Applications: Control of single video games or digital systems.

Part 2 – Theano for Deep Learning

Theano Basics

• Introduction
• Installation and Configuration

Theano Functions

• Inputs, outputs, updates, and givens

Training and Optimization of Neural Networks using Theano

• Neural Network Modeling
• Logistic Regression
• Hidden Layers
• Training a network
• Computing and Classification
• Optimization
• Log Loss

Testing the Model

Part 3 – DNN using TensorFlow

TensorFlow Basics

• Creation, initialization, saving, and restoring TensorFlow variables
• Feeding, reading, and preloading TensorFlow data
• Leveraging TensorFlow infrastructure for scalable model training
• Visualizing and evaluating models with TensorBoard

TensorFlow Mechanics

• Preparing the Data
• Downloading
• Inputs and Placeholders
• Building the Graphs
  • Inference
  • Loss
  • Training
• Training the Model
  • The Graph
  • The Session
  • Training Loop
• Evaluating the Model
  • Building the Eval Graph
  • Eval Output

The Perceptron

• Activation functions
• The perceptron learning algorithm
• Binary classification using the perceptron
• Document classification using the perceptron
• Limitations of the perceptron

From Perceptron to Support Vector Machines

• Kernels and the kernel trick
• Maximum margin classification and support vectors

Artificial Neural Networks

• Nonlinear decision boundaries
• Feedforward and feedback artificial neural networks
• Multilayer perceptrons
• Minimizing the cost function
• Forward propagation
• Back propagation
• Enhancing neural network learning capabilities

Convolutional Neural Networks

• Goals
• Model Architecture
• Principles
• Code Organization
• Launching and Training the Model
• Evaluating a Model

Brief introductions to the following modules (coverage dependent on time availability):

TensorFlow – Advanced Usage

• Threading and Queues
• Distributed TensorFlow
• Writing Documentation and Sharing your Model
• Customizing Data Readers
• Manipulating TensorFlow Model Files

TensorFlow Serving

• Introduction
• Basic Serving Tutorial
• Advanced Serving Tutorial
• Serving Inception Model Tutorial