How Smart Machines Think by Sean Gerrish (PDF)

Description

Everything you’ve always wanted to know about self-driving cars, Netflix recommendations, IBM’s Watson, and video game-playing computer programs.The future is here: Self-driving cars are on the streets, an algorithm gives you movie and TV recommendations, IBM’s Watson triumphed on Jeopardy over puny human brains, computer programs can be trained to play Atari games. But how do all these things work? In this book, Sean Gerrish offers an engaging and accessible overview of the breakthroughs in artificial intelligence and machine learning that have made today’s machines so smart.Gerrish outlines some of the key ideas that enable intelligent machines to perceive and interact with the world. He describes the software architecture that allows self-driving cars to stay on the road and to navigate crowded urban environments; the million-dollar Netflix competition for a better recommendation engine (which had an unexpected ending); and how programmers trained computers to perform certain behaviors by offering them treats, as if they were training a dog. He explains how artificial neural networks enable computers to perceive the world—and to play Atari video games better than humans. He explains Watson’s famous victory on Jeopardy, and he looks at how computers play games, describing AlphaGo and Deep Blue, which beat reigning world champions at the strategy games of Go and chess. Computers have not yet mastered everything, however; Gerrish outlines the difficulties in creating intelligent agents that can successfully play video games like StarCraft that have evaded solution—at least for now.Gerrish weaves the stories behind these breakthroughs into the narrative, introducing readers to many of the researchers involved, and keeping technical details to a minimum. Science and technology buffs will find this book an essential guide to a future in which machines can outsmart people.

User’s Reviews

Reviews from Amazon users which were colected at the time this book was published on the website:

⭐The most important thing for readers to understand about this book is that while it has somewhat been repurposed for a general audience, that’s not really who this book is for. It’s for college level Computer Science students getting their introduction into the topics of Machine Learning and Beysian programming models. Put another way, this is not a conversational book about “how those amazing computers” work; it is half way between a programming text and that, but certainly it’s “down in the weeds” a bit. Pretty dry stuff if you’re not familiar with the basic underpinnings of how computers work and basic programming concepts.That’s not to say you need to be a math wiz or a programmer to understand its basic tenets, but the more you understand computers the easier it will be. This is also not an end-to-end read. This is a “pick the topic / chapter that looks interesting” and read it a couple times to get an idea of how your favorite TV app or self-driving car concept works, without becoming an engineer yourself. There’s not much in the way of philosophical stuff, and in general like most books about “artificial intelligence,” it’s really about Machine Learning (ML) and how that works. More on this below.In short, when you feed a very powerful (many times more powerful than the one on your desk) computer a enormous amount of data of a very specific kind, and teach it to recognize very specific kinds of patterns in the data, and then run it through millions of times, the computer begins to “learn” to recognize things on its own, thus improving it’s output quality over time. This is how computers are trained to play games like Chess or Go, and then beat human competitors. It’s a false conclusion to say the computer is “thinking the game of Chess or Go” the way a person would. Instead it’s leveraging a huge amount stored knowledge of statistical patterns and applying it in the form of “moves” that are governed by a very simple set of rules. Put another way, a computer can be made to be a really awesome Go player, but if you asked the same computer why the game of Go is challenging or useful or good, it can’t do any of that. All it knows are patterns of moves and statistical probability of success of the next move, as defined by the rules.This kind of technology is not a “thinking” the way you and I think (which is what “artificial intelligence” requires), but really just learning through trial and millions of errors, then correcting those errors slowly over time, so the computer can learn to do stuff more efficiently for us, whether that’s put up recommended shows on our app, drive a car around a corner, or fix the green eyes on your pet photos without you having to do anything other than click a button.True “artificial intelligence” in a computer is a LONG WAYS away by most experts’ reckoning but this type of computer interaction and activity is a required stepping stone. So if you’re interested in learning about this one stepping stone at a deeper level, this book is for you. Otherwise there are better alternatives.

⭐This was one of the most enjoyable books I have read in a while. It presents a general overview of how various AI-powered systems work in technically accurate but lightweight, hobby-level reading terms that still give enough details to satisfy professionals in analytics, AI, and optimization.This book had the best explanation of how autonomous cars work algorithmically. Routing algorithms have been around for several decades. The hard part is recognizing pedestrians and fire hydrants and moving vehicles versus open roads in real time. Enter neural networks, which took image classification to the next level making the technology for driverless cars nearly obtainable.Autonomous vehicle algorithms is just one section of the book. Also discussed is how statistical learning can be used to power a Netflix recommendation engine. How AlphaGo works and how it involves game tree search and deep learning. How Jeopardy-playing Watson works and the natural language processing techniques that make it possible. And the challenges of building an agent-based AI that can play Starcraft.In a world of AI hype with thousands of shallow Business Insider articles, what makes this book stand out is the very knowledgeable author. The author is a machine learning PhD at Google with many years of practical, hands-on experience building model-based systems. Books by such practitioners are quite rare (since practitioners rarely take the time to write them!) and should be cherished.

⭐The book reviews the “under the hood” of several famous success stories of artificial intelligence. Some pictures are not clear in the Kindle version.

⭐Good ideas

⭐Author had tried to hard to teach machine learning in bare-bones technicality but I’m not able to grasp it at many levels.

⭐Good info

⭐O livro aborda assuntos como machine learning e traz explicações simples para descrever o funcionamento das maquinas inteligentes. Sem falar das histórias de bastidores, muito bom!

⭐

⭐A capa é só uma aparato padrão com um papel em volta, ou seja, não é impresso diretamente na capa dura.

⭐

Keywords

Free Download How Smart Machines Think in PDF format
How Smart Machines Think PDF Free Download
Download How Smart Machines Think 2018 PDF Free
How Smart Machines Think 2018 PDF Free Download
Download How Smart Machines Think PDF
Free Download Ebook How Smart Machines Think