InformIT – Machine Vision, GANs, and Deep Reinforcement Learning LiveLessons, 2nd Edition 2020-3

Published on: 2021-06-12 16:22:55

Categories: 28

Description

Machine Vision, GANs, and Deep Reinforcement Learning LiveLessons, 2nd Edition, is an intuitive introductory course for deep learning teaching that teaches three of the most popular deep learning topics. Modern machine vision has surpassed the human ability to recognize images, recognize objects, and image segmentation tasks with the introduction of automated systems. Adversarial networks, by forming two deep learning networks against each other in a forger-detective relationship, allow the creation and forgery of real-time images with objects of the user’s choice.

What you will learn in Machine Vision, GANs, and Deep Reinforcement Learning LiveLessons, 2nd Edition:

Understand high-level theory and key language around machine learning, reinforcement learning, and adversarial networks
Build state-of-the-art models for image recognition, object recognition, and image segmentation
GAN architecture that has the ability to produce compelling images in the form of human imagination
Build a deep RL assistant that can adapt to a variety of environments, such as those offered by OpenAI Gym.
Perform automated experiments to optimize deep reinforcement learning models
Understand the current limitations of artificial intelligence and how to overcome them in the future

Course specifications

Publisher: InformIT
Instructors: Jon Krohn
Language: English
Level: Average
Lessons: 45
Duration: 6 hours and 6 minutes

Course topics:

Lesson 1: Orientation
Topics
1.1 Running the Hands-On Code Examples in Jupyter Notebooks
1.2 Review of Prerequisite Deep Learning Theory
1.3 A Sneak Peak

2 Lesson: Convolutional Neural Networks for Machine Vision
Topics
2.1 Convolutional Layers
2.2 Convolutional Filter Hyperparameters
2.3 Activation Pooling and Flattening
2.4 Building A Convnet in Tensorflow
2.5 Convnet Architectures Model
2.6 Residual Networks
2.7 Image Segmentation
2.8 Object Detection
2.9 Transfer Learning
2.10 Capsule Networks

Lesson 3: Generative Adversarial Networks for Creativity
Topics
3.1 A Boozy All-Nighter
3.2 Latent Space: Arithmetic on Fake Human Faces
3.3 Style Transfer: Converting Photos into Monet (and Vice Versa)
3.4 Applications of GANs
3.5 Essential GAN Theory
3.6 The “Quick, Draw! ” Dataset
3.7 The Discriminator Network
3.8 The Generator Network
3.9 Training the Adversarial Network

Lesson 4: Deep Reinforcement Learning
Topics
4.1 Three Categories of Machine Learning Problems
4.2 When Reinforcement Learning Becomes Deep
4.3 Applications to Video Games
4.4 Applications to Board Games
4.5 Real-World Applications
4.6 Reinforcement Learning Environments
4.7 Three Categories of Artificial Intelligence

Lesson 5: Deep Q-Learning and Beyond
Topics
5.1 The Cart-Pole Game
5.2 Essential Reinforcement Learning Theory
5.3 Deep Q-Learning Networks
5.4 Defining a DQN Agent
5.5 Interacting with an Environment
5.6 Hyperparameter Optimization with SLM Lab
5.7 Agents Beyond DQN
5.8 Datasets, Project Ideas, and Resources for Self-Study
5.9 Approaching Artificial General Intelligence

Summary

Course prerequisites:

The author’s Deep Learning with TensorFlow, Keras, and PyTorch LiveLessons , or familiarity with the topics covered in Chapters 5 through 9 of his book Deep Learning Illustrated , are a prerequisite.