Oreilly – Deep Reinforcement Learning in Action, Video Edition 2020-3

Apr 18, 2025 - 18:18

0 0

Deep Reinforcement Learning in Action, Video Edition. This course teaches you how to use deep reinforcement learning to build AI agents that can learn and improve based on feedback from their environment. This technique allows algorithms to learn new skills through trial and error, just like humans. Key points: Learning through experience and feedback, Solving complex problems using AI, Building human-like AI agents, Using popular tools like PyTorch and OpenAI Gym.

What you will learn:

Building and Training DRL Networks: Learn how to build and train neural networks for deep reinforcement learning.
Popular DRL Algorithms: Introduction to the most widely used deep reinforcement learning algorithms for solving various problems.
Evolutionary Algorithms: Learn about evolutionary algorithms to foster curiosity and multi-agent learning.
Practical Implementation: All examples are presented as Jupyter notebooks so you can easily run them.

Who is it suitable for?

This course is suitable for people with intermediate knowledge of the Python programming language and deep learning.

Course details: Deep Reinforcement Learning in Action, Video Edition

Publisher: Oreilly
Instructor: Brandon Brown , Alexander Zai
Training level: Beginner to advanced
Training duration: 12 hours

Course topics

Part 1. Foundations
Chapter 1. What is reinforcement learning?
Chapter 1. Reinforcement learning
Chapter 1. Dynamic programming versus Monte Carlo
Chapter 1. The reinforcement learning framework
Chapter 1. What can I do with reinforcement learning?
Chapter 1. Why deep reinforcement learning?
Chapter 1. Our didactic tool: String diagrams
Chapter 1. What’s next?
Chapter 1. Summary
Chapter 2. Modeling reinforcement learning problems: Markov decision processes
Chapter 2. Solving the multi-arm bandit
Chapter 2. Applying bandits to optimize ad placements
Chapter 2. Building networks with PyTorch
Chapter 2. Solving contextual bandits
Chapter 2. The Markov property
Chapter 2. Predicting future rewards: Value and policy functions
Chapter 2. Summary
Chapter 3. Predicting the best states and actions: Deep Q-networks
Chapter 3. Navigating with Q-learning
Chapter 3. Preventing catastrophic forgetting: Experience replay
Chapter 3. Improving stability with a target network
Chapter 3. Review
Chapter 3. Summary
Chapter 4. Learning to pick the best policy: Policy gradient methods
Chapter 4. Reinforcing good actions: The policy gradient algorithm
Chapter 4. Working with OpenAI Gym
Chapter 4. The REINFORCE algorithm
Chapter 4. Summary
Chapter 5. Tackling more complex problems with actor-critic methods
Chapter 5. Distributed training
Chapter 5. Advantage actor-critic
Chapter 5. N-step actor-critic
Chapter 5. Summary
Part 2. Above and beyond
Chapter 6. Alternative optimization methods: Evolutionary algorithms
Chapter 6. Reinforcement learning with evolution strategies
Chapter 6. A genetic algorithm for CartPole
Chapter 6. Pros and cons of evolutionary algorithms
Chapter 6. Evolutionary algorithms as a scalable alternative
Chapter 6. Summary
Chapter 7. Distributional DQN: Getting the full story
Chapter 7. Probability and statistics revisited
Chapter 7. The Bellman equation
Chapter 7. Distributional Q-learning
Chapter 7. Comparing probability distributions
Chapter 7. Dist-DQN on simulated data
Chapter 7. Using distributional Q-learning to play Freeway
Chapter 7. Summary
Chapter 8. Curiosity-driven exploration
Chapter 8. Inverse dynamics prediction
Chapter 8. Setting up Super Mario Bros.
Chapter 8. Preprocessing and the Q-network
Chapter 8. Setting up the Q-network and policy function
Chapter 8. Intrinsic curiosity module
Chapter 8. Alternative intrinsic reward mechanisms
Chapter 8. Summary
Chapter 9. Multi-agent reinforcement learning
Chapter 9. Neighborhood Q-learning
Chapter 9. The 1D Ising model
Chapter 9. Mean field Q-learning and the 2D Ising model
Chapter 9. Mixed cooperative-competitive games
Chapter 9. Summary
Chapter 10. Interpretable reinforcement learning: Attention and relational models
Chapter 10. Relational reasoning with attention
Chapter 10. Implementing self-attention for MNIST
Chapter 10. Multi-head attention and relational DQN
Chapter 10. Double Q-learning
Chapter 10. Training and attention visualization
Chapter 10. Summary
Chapter 11. In conclusion: A review and roadmap
Chapter 11. The uncharted topics in deep reinforcement learning
Chapter 11. The end
Appendix. Mathematics, deep learning, PyTorch
Appendix. Calculus
Appendix. Deep learning
Appendix. PyTorch

Course images

Deep Reinforcement Learning in Action, Video Edition

Sample course video

Installation Guide

After Extract, view with your favorite player.

Subtitles: None

Quality: 720p

Download link

Download Part 1 – 1 GB

Download Part 2 – 233 MB

File(s) password: www.downloadly.ir

File size

1.2 GB

Oreilly – Deep Reinforcement Learning in Action, Video Edition 2020-3