
Build Intelligent Systems
Train AI Agents That Learn from ExperienceBy Nathan UnderwoodLength7h 14m
About this audiobook
Are you ready to build intelligent systems from the ground up? This practical guide takes you from search algorithms and logic to machine learning and neural networks, with step-by-step projects that turn theory into working code. Whether you're a beginner or a pro, you'll gain the skills to create real AI applications. Unlike The Other Wes Moore: One Name, Two Fates (One World Essentials) or The Other Emily, this book is your hands-on path to mastering AI—no fluff, just results.
What You'll Build
Search engines and game-playing agents
Logic-based reasoning systems
Machine learning models for prediction
Neural networks for image and text
End-to-end intelligent systems
Each chapter includes code examples, exercises, and a final project that ties everything together. You'll learn by doing, not just reading. Nathan Underwood's approach ensures you understand the 'why' behind each algorithm, so you can adapt and innovate.
Why This Book?
Clear explanations for complex topics
Real-world examples and datasets
Progressive difficulty from basics to advanced
Focus on building, not just theory
Start your journey today and create systems that think, learn, and adapt. Perfect for students, developers, and AI enthusiasts.
This hands-on Other guide is written to be used at the keyboard: every concept is paired with something you can run, adapt, and keep. You move from first principles to real, working results, with the common errors and fixes called out along the way so you are never stuck for long.
Audiobook details
GenreTechnology
Length7 hrs 14 mins
Narrated byListen with 1,000+ voices
FormateBook with Audio
LanguageEnglish
Table of contents
1Preface
2Chapter 1 — Your First AI Agent in Ten Lines
3Chapter 2 — The Grid World: Defining the Environment Your Agent Will Master
4Chapter 3 — State Representation: How the Agent Sees the World
5Chapter 4 — Reward Engineering: Shaping Behavior Without Hard-Coding Rules
Show all chaptersShow less
6Chapter 5 — Tabular Q-Learning: The Agent Learns Its First Policy
7Chapter 6 — Policy Evaluation and Improvement: How to Know the Agent Is Getting Smarter
8Chapter 7 — Scaling Up with Function Approximation: Neural Networks Replace the Table
9Chapter 8 — Experience Replay: Learning from Past Mistakes
10Chapter 9 — Target Networks: Preventing the Moving Target Problem
11Chapter 10 — Double DQN and Dueling Networks: Two Proven Enhancements
12Chapter 11 — Policy Gradients: Learning Actions Directly
13Chapter 12 — Actor-Critic: Combining Value and Policy for Faster Learning
14Chapter 13 — Hyperparameter Tuning: Finding the Sweet Spot
15Chapter 14 — Multi-Environment Training: Speeding Up with Parallel Agents
16Chapter 15 — Saving, Loading, and Resuming Training
17Chapter 16 — Deploying the Agent as a Web App
18Chapter 17 — Testing and Debugging in Production
19Chapter 18 — Extending to New Environments: From Grid World to Real-World Problems
20About the Author