Название: The AI Model Handbook: A guide to the world of artificial intelligence modeling
Автор: Minh Trinh
Издательство: Rodeo Press
Год: 2021
Страниц: 358
Язык: английский
Формат: pdf, epub
Размер: 22.1 MB

This book introduces in a non-technical way Artificial Intelligence (AI), Machine Learning, and the most common models used in production. It covers supervised and unsupervised learning, Deep Learning, natural language processing (NLP), computer vision, generative adversarial networks, graph neural networks, recommender systems, and causal inference.

Machine learning studies how computer programs can learn with experience (data) to make better forecasts and decisions. Learning can involve actual data or simulated data. We can see machine learning as a subcategory of artificial intelligence. It is unclear whether machine learning is sufficient to “solve intelligence.” In some specific contexts, though a machine can indeed be trained to outperform humans, such as in the game of Chess or Go, a more versatile intelligence still seems out of reach. The recent progress of machine learning and deep learning, in particular, has been spectacular and sees no sign of slowing down.

We can divide Machine Learning into different categories: Supervised learning, unsupervised learning, self-supervised learning, and reinforcement learning. The distinctions between these categories are sometimes not definitive. For instance, reinforcement learning can be interpreted as supervised learning since the reward can also be seen as a target. We can consider self-supervised learning as supervised learning because it uses self-contained target examples.

Who Should Read This Book?
The Artificial Intelligence Handbook Series will help you adopt AI to create a long-term sustainable competitive advantage for your organization. The AI Model Handbook should be of interest to readers who want to have a detailed knowledge of the state of machine learning without going too deep into the mathematical details of the field.

Outline of This Book
In Chapter 1, we introduce artificial intelligence and machine learning. Chapters 2 and 3 go over supervised and unsupervised machine learning. Chapter 4 covers deep learning that has revolutionized machine learning in the past ten years. Chapter 5 is the most technical and discusses reinforcement learning, the technique widely used at DeepMind to design intelligent agents. Chapters 6 and 7 are more applied and focus on Natural Language Processing and Computer Vision. Chapter 8 covers Generative Adversarial Networks that can generate new realistic data. Chapter 9 introduces Graphical Neural Networks, Chapter 10, Recommender Systems, and Chapter 11, Causal Inference in Machine Learning. Each chapter starts with a personal quote and the portraits of some eminent AI scientists. More biographical information is available in the appendix.