Book Image

Artificial Intelligence By Example - Second Edition

By : Denis Rothman
Book Image

Artificial Intelligence By Example - Second Edition

By: Denis Rothman

Overview of this book

AI has the potential to replicate humans in every field. Artificial Intelligence By Example, Second Edition serves as a starting point for you to understand how AI is built, with the help of intriguing and exciting examples. This book will make you an adaptive thinker and help you apply concepts to real-world scenarios. Using some of the most interesting AI examples, right from computer programs such as a simple chess engine to cognitive chatbots, you will learn how to tackle the machine you are competing with. You will study some of the most advanced machine learning models, understand how to apply AI to blockchain and Internet of Things (IoT), and develop emotional quotient in chatbots using neural networks such as recurrent neural networks (RNNs) and convolutional neural networks (CNNs). This edition also has new examples for hybrid neural networks, combining reinforcement learning (RL) and deep learning (DL), chained algorithms, combining unsupervised learning with decision trees, random forests, combining DL and genetic algorithms, conversational user interfaces (CUI) for chatbots, neuromorphic computing, and quantum computing. By the end of this book, you will understand the fundamentals of AI and have worked through a number of examples that will help you develop your AI solutions.
Table of Contents (23 chapters)
21
Other Books You May Enjoy
22
Index

Chapter 13 – Visualizing Networks with TensorFlow 2.x and TensorBoard

  1. A CNN always has the same number of layers. (Yes | No)

    No. A CNN does not have the same number of layers or even the same type of layers. The number of layers is part of the work to optimize an artificial neural network.

  2. ReLU is the best activation function. (Yes | No)

    No. ReLU is an efficient activation function, but there are others such as leaky ReLU, softmax, sigmoid, and tanh.

  3. It is not necessary to compile a sequential classifier. (Yes | No)

    No. The assertion should be yes – it is necessary.

  4. The output of a layer is best viewed without running a prediction. (Yes | No)

    No. The output of a layer and a prediction are unrelated. The output of the layer can be the transformation of a layer (convolutional, pooling, dropout, flattening, other) or a prediction.

  5. The names of the layers mean nothing when viewing...