MEAP Edition  Manning Early Access Program  Deep Learning with R  Version 1        Copyright 2017 Manning Publications      For more information on this and other Manning titles go to   www.manning.com  ©Manning Publications Co. We welcome reader comments about anything in the manuscript - other than typos and other simple mistakes. These will be cleaned up during production of the book by copyeditors and proofreaders. https://forums.manning.com/forums/deep-learning-with-r Licensed to <null> welcome Thank you for purchasing the MEAP for Deep Learning with R. If you are looking for a resource to learn about deep learning from scratch and to quickly become able to use this knowledge to solve real-world problems, you have found the right book. Deep Learning with R is meant for statisticians, analysts, engineers and students with a reasonable amount of R experience, but no significant knowledge of machine learning and deep learning. This book is an adaptation of my previously published Deep Learning with Python, with all of the code examples using the R interface to Keras. The goal of the book is to provide a learning resource for the R community that goes all the way from basic theory to advanced practical applications. Deep learning is an immensely rich subfield of machine learning, with powerful applications ranging from machine perception to natural language processing, all the way up to creative AI. Yet, its core concepts are in fact very simple. Deep learning is often presented as shrouded in a certain mystique, with references to algorithms that “work like the brain”, that “think” or “understand”. Reality is however quite far from this science- fiction dream, and I will do my best in these pages to dispel these illusions. I believe that there are no difficult ideas in deep learning, and that’s why I started this book, based on premise that all of the important concepts and applications in this field could be taught to anyone, with very few prerequisites. This book is structured around a series of practical code examples, demonstrating on real- world problems every the notions that gets introduced. I strongly believe in the value of teaching using concrete examples, anchoring theoretical ideas into actual results and tangible code patterns. These examples all rely on Keras, the deep learning library. When I released the initial version of Keras almost two years ago, little did I know that it would quickly skyrocket to become one of the most widely used deep learning frameworks. A big part of that success is that Keras has always put ease of use and accessibility front and center. This same reason is what makes Keras a great library to get started with deep learning, and thus a great fit for this book. By the time you reach the end of this book, you will have become a Keras expert. I hope that you will this book valuable —deep learning will definitely open up new intellectual perspectives for you, and in fact it even has the potential to transform your career, being the most in-demand scientific specialization these days. I am looking forward to your reviews and comments. Your feedback is essential in order to write the best possible book, that will benefit the greatest number of people. — François Chollet ©Manning Publications Co. We welcome reader comments about anything in the manuscript - other than typos and other simple mistakes. These will be cleaned up during production of the book by copyeditors and proofreaders. https://forums.manning.com/forums/deep-learning-with-r Licensed to <null> brief contents PART 1: INTRODUCTION TO DEEP LEARNING 1 What is deep learning? 2 Before we begin: the mathematical blocks of neural networks 3 Getting started with neural networks 4 Fundamentals of machine learning PART 2: DEEP LEARNING IN PRACTICE 5 Deep learning for computer vision 6 Deep learning for text and sequences 7 Advanced deep learning best practices 8 Generative deep learning 9 Conclusions APPENDIXES: A Installing Keras and its dependencies on Ubuntu B Running RStudio Server on a EC2 GPU instance ©Manning Publications Co. We welcome reader comments about anything in the manuscript - other than typos and other simple mistakes. These will be cleaned up during production of the book by copyeditors and proofreaders. https://forums.manning.com/forums/deep-learning-with-r Licensed to <null> 1 1 What is deep learning? This chapter covers High-level definitions of fundamental concepts Timeline of the development of machine learning Key factors behind deep learning’s rising popularity and future potential In the past few years, artificial intelligence (AI) has been a subject of intense media hype. Machine learning, deep learning, and AI come up in countless articles, often outside of technology-minded publications. We’re being promised a future of intelligent chatbots, self-driving cars, and virtual assistants—a future sometimes painted in a grim light and other times as an utopia, where human jobs would be scarce and most economic activity would be handled by robots or AI agents. This chapter provides essential context around artificial intelligence, machine learning, and deep learning. 1.1 Artificial intelligence, machine learning, and deep learning As a future or current practitioner of machine learning, it’s important to be able to recognize the signal in the noise so that you can tell world-changing developments from mere overhyped press releases. Our future is at stake, and it’s a future in which you have an active role to play: after reading this book, you’ll be one of those who develop the AIs. So let’s tackle these questions: What has deep learning achieved so far? How significant is it? Where are we headed next? Should you believe the hype? First, we need to define clearly what we’re talking about when we mention AI. What are artificial intelligence, machine learning, and deep learning (see figure 1.1)? How do they relate to each other? ©Manning Publications Co. We welcome reader comments about anything in the manuscript - other than typos and other simple mistakes. These will be cleaned up during production of the book by copyeditors and proofreaders. https://forums.manning.com/forums/deep-learning-with-r Licensed to <null> 2 Figure 1.1 Artificial intelligence, machine learning, and deep learning 1.1.1 Artificial intelligence Artificial intelligence was born in the 1950s, when a handful of pioneers from the nascent field of computer science started asking whether computers could be made to "think"—a question whose ramifications we’re still exploring today. A concise definition of the field would be as follows: the effort to automate intellectual tasks normally performed by humans. As such, AI is a general field that encompasses machine learning and deep learning, but that also includes many more approaches that don’t involve any learning. Early chess programs, for instance, only involved hard-coded rules crafted by programmers, and didn’t qualify as machine learning. For a fairly long time, many experts believed that human-level artificial intelligence could be achieved by having programmers handcraft a sufficiently large set of explicit rules for manipulating knowledge. This approach is known as symbolic AI, and it was the dominant paradigm in AI from the 1950s to the late 1980s. It reached its peak popularity during the expert systems boom of the 1980s. Although symbolic AI proved suitable to solve well-defined, logical problems, such as playing chess, it turned out to be intractable to figure out explicit rules for solving more complex, fuzzy problems, such as image classification, speech recognition, and language translation. A new approach arose to take symbolic AI’s place: machine learning. ©Manning Publications Co. We welcome reader comments about anything in the manuscript - other than typos and other simple mistakes. These will be cleaned up during production of the book by copyeditors and proofreaders. https://forums.manning.com/forums/deep-learning-with-r Licensed to <null> 3 1.1.2 Machine Learning In Victorian England, Lady Ada Lovelace was a friend and collaborator of Charles Babbage, the inventor of the Analytical Engine: the first known general-purpose, mechanical computer. Although visionary and far ahead of its time, the Analytical Engine wasn’t meant as a general-purpose computer when it was designed in the 1830s and 1840s, because the concept of general-purpose computation was yet to be invented. It was merely meant as a way to use mechanical operations to automate certain computations from the field of mathematical analysis—hence the name Analytical Engine. In 1843, Ada Lovelace remarked on the invention, "The Analytical Engine has no pretensions whatever to originate anything. It can do whatever we know how to order it to perform.… Its province is to assist us in making available what we’re already acquainted with." This remark was later quoted by AI pioneer Alan Turing as "Lady Lovelace’s objection" in his landmark 1950 paper "Computing Machinery and Intelligence,"1 which introduced the Turing test as well as key concepts that would come to shape AI. Turing was quoting Ada Lovelace while pondering whether general-purpose computers could be capable of learning and originality, and he came to the conclusion that they could. Footnote 1mA. M. Turing, "Computing Machinery and Intelligence," Mind 59, no. 236 (1950): 433-460. Machine learning arises from this question: could a computer go beyond "what we know how to order it to perform" and learn on its own how to perform a specified task? Could a computer surprise us? Rather than programmers crafting data-processing rules by hand, could a computer automatically learn these rules by looking at data? This question opens the door to a new programming paradigm. In classical programming, the paradigm of symbolic AI, humans input rules (a program) and data to be processed according to these rules, and out come answers (see figure 1.2). With machine learning, humans input data as well as the answers expected from the data, and out come the rules. These rules can then be applied to new data to produce original answers. Figure 1.2 Machine learning: a new programming paradigm A machine-learning system is trained rather than explicitly programmed. It’s presented with many examples relevant to a task, and it finds statistical structure in these examples that eventually allows the system to come up with rules for automating the task. For instance, if you wished to automate the task of tagging your vacation pictures, ©Manning Publications Co. We welcome reader comments about anything in the manuscript - other than typos and other simple mistakes. These will be cleaned up during production of the book by copyeditors and proofreaders. https://forums.manning.com/forums/deep-learning-with-r Licensed to <null> 4 you could present a machine-learning system with many examples of pictures already tagged by humans, and the system would learn statistical rules for associating specific pictures to specific tags. Although machine learning only started to flourish in the 1990s, it has quickly become the most popular and most successful subfield of AI, a trend driven by the availability of faster hardware and larger datasets. Machine learning is tightly related to mathematical statistics, but it differs from statistics in several important ways. Unlike statistics, machine learning tends to deal with large, complex datasets (such as a dataset of millions of images, each consisting of tens of thousands of pixels) for which classical statistical analysis such as Bayesian analysis would be impractical. As a result, machine learning, and especially deep learning, exhibits comparatively little mathematical theory—maybe too little—and is engineering oriented. It’s a hands-on discipline in which ideas are proven empirically much more often than theoretically. 1.1.3 Learning representations from data To define deep learning and understand the difference between deep learning and other machine-learning approaches, first we need some idea of what machine-learning algorithms do. We just stated that machine learning discovers rules to execute a data-processing task, given examples of what’s expected. So, to do machine learning, we need three things: Input data points—For instance, if the task is speech recognition, these data points could be sound files of people speaking. If the task is image tagging, they could be picture files. Examples of the expected output—In a speech-recognition task, these could be human-generated transcripts of sound files. In an image task, expected outputs could tags such as "dog", "cat", and so on. A way to measure whether the algorithm is doing a good job—This is necessary in order to determine the distance between the algorithm’s current output and its expected output. The measurement is used as a feedback signal to adjust the way the algorithm works. This adjustment step is what we call learning. A machine-learning model transforms its input data into meaningful output, a process that is "learned" from exposure to known examples of inputs and outputs. Therefore, the central problem in machine learning and deep learning is to meaningfully transform data: in other words, to learn useful representations of the input data at hand—representations that get us closer to the expected output. Before we go any further: what’s a representation? At its core, it’s a different way to look at data—to represent or encode data. For instance, a color image can be encoded in the RGB format (red-green-blue) or in the HSV format (hue-saturation-value): these are two different representations of the same data. Some tasks that may be difficult with one representation can become easy with another. For example, the task "select all red pixels in the image" is simpler in the RBG format, whereas "make the image less saturated" is simpler in the HSV format. Machine-learning models are all about finding appropriate representations for their input data—transformations of the data that make it more amenable to the task at hand, such as a classification task. ©Manning Publications Co. We welcome reader comments about anything in the manuscript - other than typos and other simple mistakes. These will be cleaned up during production of the book by copyeditors and proofreaders. https://forums.manning.com/forums/deep-learning-with-r Licensed to <null> 5 Let’s make this concrete. Consider an x axis, a y axis, and some points represented by their coordinates in the (x, y) system, as shown in figure 1.3. Figure 1.3 Some sample data As you can see, we have a few white points and a few black points. Let’s say we want to develop an algorithm that can take the coordinates (x, y) of a point and output whether that point is likely to be black or to be white. In this case, The inputs are the coordinates of our points. The expected outputs are the colors of our points. A way to measure whether our algorithm is doing a good job could be, for instance, the percentage of points that are being correctly classified. What we need here is a new representation of our data that cleanly separates the white points from the black points. One transformation we could use, among many other possibilities, would be a coordinate change, illustrated in figure 1.4. Figure 1.4 Coordinate change In this new coordinate system, the coordinates of our points can be said to be a new representation of our data. And it’s a good one! With this representation, the black/white classification problem can be expressed as a simple rule: black points are such that x 0 or "white points are such that x < 0". This new representation basically solves the classification problem. In this case, we defined the coordinate change by hand. But if instead we tried systematically searching for different possible coordinate changes, and used as feedback ©Manning Publications Co. We welcome reader comments about anything in the manuscript - other than typos and other simple mistakes. These will be cleaned up during production of the book by copyeditors and proofreaders. https://forums.manning.com/forums/deep-learning-with-r Licensed to <null> 6 the percentage of points being correctly classified, then we would be doing machine learning. Learning, in the context of machine learning, describes an automatic search process for better representations. All machine-learning algorithms consist of automatically finding such transformations that turn data into more useful representations for a given task. These operations can be coordinate changes, as you just saw, or linear projections (which may destroy information), translations, nonlinear operations (such as select all points such that x 0), and so on. Machine-learning algorithms aren’t usually creative in finding these transformations; they’re merely searching through a predefined set of operations, called a hypothesis space. So that’s what machine learning is, technically: searching for useful representations of some input data, within a predefined space of possibilities, using guidance from some feedback signal. This simple idea allows for solving a remarkably broad range of intellectual tasks, from speech recognition to autonomous car driving. Now that you understand what we mean by learning, let’s take a look at what makes deep learning special. 1.1.4 The "deep" in deep learning Deep learning is a specific subfield of machine learning: a new take on learning representations from data that puts an emphasis on learning successive layers of increasingly meaningful representations. The deep in deep learning isn’t a reference to any kind of deeper understanding achieved by the approach; rather, it stands for this idea of successive layers of representations. How many layers contribute to a model of the data is called the depth of the model. Other appropriate names for the field could have been layered representations learning and hierarchical representations learning. Modern deep learning often involves tens or even hundreds of successive layers of representation—and they’re all learned automatically from exposure to training data. Meanwhile, other approaches to machine learning tend to focus on learning only one or two layers of representation of the data; hence they’re sometimes called shallow learning . In deep learning, these layered representations are (almost always) learned via models called neural networks, structured in literal layers stacked one after the other. The term neural network is a reference to neurobiology, but although some of the central concepts in deep learning were developed in part by drawing inspiration from our understanding of the brain, deep learning models are not models of the brain. There’s no evidence that the brain implements anything like the learning mechanisms used in modern deep-learning models. You may come across pop-science articles proclaiming that deep learning works like the brain or was modeled after the brain, but that isn’t the case. It would be confusing and counterproductive for newcomers to the field to think of deep learning as being in any way related to the neurobiology; you don’t need that shroud of "just like our ©Manning Publications Co. We welcome reader comments about anything in the manuscript - other than typos and other simple mistakes. These will be cleaned up during production of the book by copyeditors and proofreaders. https://forums.manning.com/forums/deep-learning-with-r Licensed to <null>