Introduction
With advances in AI, increasing availability of public data, and faster, more powerful computers, interest in deep learning is growing every day. Using artificial neural networks, this subfield of machine learning has made amazing progress in areas like image recognition, natural language processing, and identifying complex patterns in data. It’s this technology that has made possible things like virtual personal assistants, interactive chatbots, self-driving cars, robotics-assisted surgery, and life-saving medical diagnoses that other methods might have missed. Indeed, the applications of this world-changing technology are virtually limitless.
For this sampler, I chose chapters from three Manning books that explore the basics of deep learning. In the chapter “What is Deep Learning?” from Deep Learning with Python, author and deep learning researcher François Chollet offers a bird’s eye view of fundamental deep learning concepts and spotlights the factors that are integral to DL’s ever-increasing popularity and future potential.
A chapter called “Fundamental concepts: how do machines learn?” from my own book, Grokking Deep Learning, explains how deep learning, machine learning, and AI relate to each other. You’ll also take a look at supervised vs. unsupervised learning as well as parametric vs. non-parametric learning.
TensorFlow, created by the Google Brain team, is a highly popular open source library for large-scale machine learning. In “Sequence-to-sequence models for chatbots,” a chapter from Machine Learning with TensorFlow by Nishant Shukla with Kenneth Fricklas, both seasoned machine learning practitioners, you’ll implement TensorFlow as you build a chatbot by feeding sequences into a neural network. As you work, you’ll explore classification, recurrent neural networks (RNNs), seq2seq architecture, vector representation, and embedding natural language in TensorFlow.
I hope this collection of chapters featuring deep learning basics inspires you as you venture into this rapidly advancing field.