Pro Deep Learning with TensorFlow: A Mathematical Approach to Advanced Artificial Intelligence in Python

by Santanu Pattanayak

Deploy deep learning solutions in production with ease using TensorFlow. You'll also develop the mathematical understanding and intuition required to invent new deep learning architectures and solutions on your own.Pro Deep Learning with TensorFlow provides practical, hands-on expertise so you can learn deep learning from scratch and deploy meaningful deep learning solutions. This book will allow you to get up to speed quickly using TensorFlow and to optimize different deep learning architectures.All of the practical aspects of deep learning that are relevant in any industry are emphasized in this book. You will be able to use the prototypes demonstrated to build new deep learning applications. The code presented in the book is available in the form of iPython notebooks and scripts which allow you to try out examples and extend them in interesting ways.You will be equipped with the mathematical foundation and scientific knowledge to pursue research in this field and give back to the community. What You'll LearnUnderstand full stack deep learning using TensorFlow and gain a solid mathematical foundation for deep learningDeploy complex deep learning solutions in production using TensorFlowCarry out research on deep learning and perform experiments using TensorFlowWho This Book Is ForData scientists and machine learning professionals, software developers, graduate students, and open source enthusiasts

Genres: Programming

730 pages, Kindle Edition

Published December 6, 2017

Reviews

[Hồ Vinh · Rating 5 out of 5]

A very good introductory book for Deep Learning beginner: the math is just right, plenty of intuitive explanations, and it tackles many fundamental concepts that may take months to understand if you purely study from original papers. I can assure you this is a high-quality book, though admittedly, sometimes there are pages full of Python code that may signal a wrong impression. Note that, the book was written in 2017, and three years is a big gap given the community's growth pace, so you would not expect off-the-self models here, say, BERT or GAN variations.

Here are selected topics that I think may benefit readers:
- DL-relevant topic in Linear Algebra (eigenvectors), Calculus (gradient computation, a closed-form solution to find maxima/minima of functions, Taylor series), and Probability (distributions, MLE).
- Conventional optimization techniques: an excellent section on Gradient Descent and its variant/characteristics; on par with are how to define an objective function with inequality constraints.
- Formulate and derive the solution for PCA and linear regression with regularization,
- Perceptron, details to apply Backpropagation on Neural Network. [Important]: characteristics of cost functions in high-dimensional space to navigate model updating effectively.
- Convolutional Neural Network: 2D convolution operation, image-processing filter, translation equivariance, translation invariance, DropOut, BatchNorm, transfer learning, LeNet up to Inception V3.
- Recurrent Neural Network: vector representation of words (CBOW, Skip-Gram, GloVe), Vanishing and Exploding Gradient problem, RNN, LSTM, GRU.
- Another excellent section on Restricted Boltzmann Machines with a nice touch on MCMC methods. Other unsupervised methods are Autoencoders, PCA, and ZCA Whitening.
- Introduces CV challenging tasks: Image segmentation, Image classification and localization, Object detection, and Generative Adversarial Network.