
Fundamentals of Deep Learning(2017) Stanford textbook 评分:
This booked is aimed an audience with a basic operating understanding of calculus, matrices, and Python programming. Approaching this material without this back‐ ground is possible, but likely to be more challenging. Background in linear algebra may also be helpful in navigating certain sections of
Contents Preface ····· The neural network Building intelligent Machines The Limits of Traditional Computer Programs The Mechanics of Machine Learning The Neuron Expressing linear perceptrons as neurons 23789 FeedForward Neural Networks Linear Neurons and Their limitations 12 Sigmoid, Tanh, and ReLU Neurons 13 Softmax Output layers 15 Looking forward 15 2. Training FeedForward Neural Networks 17 The FastFood Problem 17 Gradient descent The Delta Rule and Learning rates 21 Gradient Descent with Sigmoidal Neurons 22 The Backpropagation Algorithm 23 Stochastic and minibatch gradient descent 25 Test Sets, Validation Sets, and Overfitting 27 Preventing Overfitting in Deep Neural Networks 34 Summar y 37 3. Implementing Neural Networks in TensorFlow 39 What is tensor flow? How Does Tensor Flow Compare to Alternatives? Installing Tensor Flow 41 Creating and manipulating Tensor Flow variables Tensor Flow Operations 45 Placeholder tensors 45 Sessions in Tensor flow 46 Navigating Variable Scopes and Sharing variables 48 Managing models over the cPU and gPu 51 Specifying the Logistic Regression Model in Tensor Flow 52 Logging and Training the logistic Regression Model 55 Leveraging Tensor Board to Visualize Computation Graphs and Learning 58 Building a multilayer Model for MNISt in TensorFlow 59 Summary 62 4. Beyond Gradient Descent. The Challenges with Gradient Descent 63 Local Minima in the Error Surfaces of Deep Networks 64 Model identifiability 65 How Pesky are Spurious Local Minima in Deep Networks? Flat Regions in the Error Surface When the gradient Points in the Wrong Direction MomentumBased optimization a Brief view of SecondOrder methods Learning Rate Adaptation 78 Ada gradAccumulating Historical Gradients RMSPropExponentially Weighted Moving Average of Gradients AdamCombining Momentum and RMSProp 81 The Philosophy behind optimizer Selection 83 Summary 83 5. Convolutional neural networks 85 eurons in humanⅤ ISion 85 The shortcomings of Feature Selection 86 Vanilla deep neural Networks Dont Scale Filters and Feature maps Full Description of the Convolutional layer 95 Max Pooling 98 Full Architectural Description of Convolution Networks Closing the Loop on mNiSt with Convolutional Networks 101 Image Preprocessing Pipelines Enable More Robust Models 103 Accelerating Training with Batch Normalization 104 Building a Convolutional Network for CIFAR10 107 Visualizing learning in Convolutional Networks 109 Leveraging Convolutional Filters to Replicate Artistic Styles 113 Learning Convolutional Filters for Other Problem Domains 114 Summary 6. Embedding and Representation Learning................... 117 Learning lowerDimensional representations Principal component analysis 118 Motivating the autoencoder Architecture 120 Implementing an autoencoder in TensorFlow 121 Denoising to Force Robust Representations 134 Sparsity in autoencoders 137 When Context Is More Informative than the Input Vector 140 The Word2vec framework 143 Implementing the SkipGram Architecture 146 Summary 152 7. Models for Sequence analysis............. 153 Analyzing VariableLength Inputs 153 Tackling seq2seg with Neural Ngrams 155 Implementing a PartofSpeech Tagger Dependency Parsing and SyntaxNet 164 Beam Search and Global normalization 168 A Case for Stateful Deep learning models 72 Recurrent Neural Networks 173 The Challenges with Vanishing Gradients 176 Long Short Term Memory(LstM) Units 178 Tensor flow Primitives for rnn models 183 Implementing a sentiment analysis model 185 Solving seq2seq tasks with Recurrent Neural Networks 189 Augmenting recurrent Networks with Attention 191 Dissecting a Neural Translation Network 194 Summary 217 8. Memory augmented Neural Networks.................... 219 Neural turing machines 219 AttentionBased Memory access 221 NTM Memory Addressing Mechanisms 223 Differentiable Neural Computers 226 InterferenceFree Writing in DNCs 229 DNC Memory reuse 230 Temporal linking of DNC Writes 231 Understanding the dnc read head 232 The DNC Controller Network 232 Visualizing the DNC in Action 234 Implementing the dNC in Tensor Flow 237 Teaching a dnc to Read and Comprehend 242 S ummar 244 9. Deep Reinforcement Learning Deep Reinforcement Learning Masters Atari Games 245 What Is Reinforcement Learning 247 Markov Decision Processes(MDp) 248 Policy 249 Future return 250 Discounted future return 251 ExploreⅤ Exploit 251 Policy Versus Value Learning 253 Policy learning via Policy gradients 254 PoleCart with Policy Gradients 254 Openal gym 254 Creating n agent 255 Building the Model and Optimizer 257 Sampling Actions 257 Keeping Track of History 257 Policy gradient main Function 258 PGAgent Performance on PoleCart 260 QLearning and Deep QNetworks 261 The Bellman equation 261 Issues with Value Iteration 262 Approximating the QFunction 262 Deep QNetwork(DQN 263 Training DQN 263 Learning stability 263 Target QNetwork 264 Experience replay 264 From QFunction to Policy 264 DQN and the markov assumption 265 DONS Solution to the markov assumption 265 Playing breakout wth DQN 265 Building Our Architecture 268 Stackin g F1 rames 268 Setting Up Training Operations 268 Updating Our Target QNetwork 269 Implementing experience replay 269 DQN Main loop 270 DQNAgent Results on Breakout 272 Improving and Moving Beyond DQN 273 Deep recurrent QNetworks ( DRQN 273 Asynchronous Advantage ActorCritic Agent(A3C 274 UNsupervised REinforcement and Auxiliary Learning (UNREAl 275 Summary 276 Index 277 Preface With the reinvigoration of neural networks in the 2000s, deep learning has become an extremely active area of research that is paving the way for modern machine learn ing. This book uses exposition and examples to help you understand major concepts in this complicated field. Large companies such as Google, Microsoft, and Facebook have taken notice and are actively growing inhouse deep learning teams. For the rest of us, deep learning is still a pretty complex and difficult subject to grasp. Research papers are filled to the brim with jargon, and scattered online tutorials do little to help build a strong intuition for why and how deep learning practitioners approach problems. Our goal is to bridge this gap Prerequisites and objectives This booked is aimed an audience with a basic operating understanding of calculus, matrices,and Python programming. Approaching this material without this back ground is possible, but likely to be more challenging. Background in linear algebra may also be helpful in navigating certain sections of mathematical exposition By the end of the book, we hope that our readers will be left with an intuition for how to approach problems using deep learning, the historical context for modern deep learning approaches, and a familiarity with implementing deep learning algorithms using the Tensor Flow open source library Conventions Used in this book The following typographical conventions are used in this book italic Indicates new terms, URLs. email addresses filenames, and file extensions Constant width Used for program listings, as well as within paragraphs to refer to program ele ments such as variable or function names, databases data types, environment variables, statements, and keywords Constant width bold Shows commands or other text that should be typed literally by the user. Constant width italic Shows text that should be replaced with usersupplied values or by values deter mined by context Using Code Examples Supplemental material (code examples, exercises, etc. ) is available for download at https://github.com/darksigma/fundamentalsofdeeplearniNgbook This book is here to help you get your job done. In general, if example code is offered with this book, you may use it in your programs and documentation. You do not need to contact us for permission unless you're reproducing a significant portion of the code. For example, writing a program that uses several chunks of code from this book does not require permission. Selling or distributing a CDROM of examples from OReilly books does require permission. Answering a question by citing this book and quoting example code does not require permission. Incorporating a signifi cant amount of example code from this book into your product's documentation does require permission We appreciate, but do not require, attribution. An attribution usually includes the itle, author, publisher, and ISBN. For example: Fundamentals of Deep Learning by Nikhil Buduma and Nicholas Locascio(OReilly). Copyright 2017 Nikhil Buduma and nicholas locascio 9781491925614 If you feel your use of code examples falls outside fair use or the permission given above,feelfreetocontactusatpermissions@oreilly.com Safari books online SAfari Safari Books Online is an ondemand digital library that deliv ers expert content in both book and video form from the worlds leading authors in technology and business Technology professionals, software developers, web designers, and business and crea tive professionals use Safari Books Online as their primary resource for research, problem solving, learning, and certification training Safari Books Online offers a range of plans and pricing for enterprise, government, education and individuals Members have access to thousands of books, training videos, and prepublication manuscripts in one fully searchable database from publishers like O Reilly Media, Prentice Hall Professional, AddisonWesley Professional, Microsoft Press, Sams, Que Peachpit PresS, Focal Press, Cisco Press, John Wiley Sons, Syngress, Morgan Kauf mann,IBM Redbooks, Packt, Adobe Press, FT Press, Apress, Manning, New Riders, McGrawHill, Jones Bartlett, Course Technology, and hundreds more. For more information about Safari Books Online, please visit us online How to contact us Please address comments and questions concerning this book to the publisher O'Reilly media, Inc 1005 Gravenstein Highway north Sebastopol, CA95472 8009989938(in the United States or Canada) 7078290515 (international or local) 7078290104(fax) To comment or ask technical questions about this book, send email to bookques tions@oreilly.com For more information about our books courses, conferences and news, see our web siteathttp://www.oreilly.com FindusonFacebookhttp:facebookcom/oreilly FollowusonTwitterhttp://twitter.com/oreillymedia WatchusonYoutube:http://www.youtubecom/oreillymedia20180303 上传 大小：5.08MB

Fundamentals of Deep Learning 深度学习基础英文版 pdf电子书+代码
Fundamentals of Deep Learning 深度学习基础英文版 pdf电子书+代码
立即下载

Fundamentals of Deep Learning
Fundamentals of Deep Learning_ Designing NextGeneration Machine Intelligence Algorithms
立即下载

Fundamentals of Deep Learning完整非扫描版本2017
Fundamentals of Deep Learning 完整非扫描版本, 作者会不断完善更新这本书， 现在是2017年版本 Nikhil Buduma and Nicholas Locascio
立即下载

Fundamentals of Deep Learning 英文原版
Fundamentals of Deep LearningDesigning NextGeneration Machine [英文原版]
立即下载