What this book covers

Chapter 1, Getting Started with Next-Generation Artificial Intelligence through Reinforcement Learning, covers reinforcement learning through the Bellman equation based on the MDP. A case study describes how to solve a delivery route problem with a human driver and a self-driving vehicle. This chapter shows how to build an MDP from scratch in Python.

Chapter 2, Building a Reward Matrix – Designing Your Datasets, demonstrates the architecture of neural networks starting with the McCulloch-Pitts neuron. The case study describes how to use a neural network to build the reward matrix used by the Bellman equation in a warehouse environment. The process will be developed in Python using logistic, softmax, and one-hot functions.

Chapter 3, Machine Intelligence – Evaluation Functions and Numerical Convergence, shows how machine evaluation capacities have exceeded human decision-making. The case study describes a chess position and how to apply the results of an AI program to decision-making priorities. An introduction to decision trees in Python shows how to manage decision-making processes.

Chapter 4, Optimizing Your Solutions with K-Means Clustering, covers a k-means clustering program with Lloyd's algorithm and how to apply it to the optimization of automatic guided vehicles. The k-means clustering program's model will be trained and saved.

Chapter 5, How to Use Decision Trees to Enhance K-Means Clustering, begins with unsupervised learning with k-means clustering. The output of the k-means clustering algorithm will provide the labels for the supervised decision tree algorithm. Random forests will be introduced.

Chapter 6, Innovating AI with Google Translate, explains the difference between a revolutionary innovation and a disruptive innovation. Google Translate will be described and enhanced with an innovative k-nearest neighbors-based Python program.

Chapter 7, Optimizing Blockchains with Naive Bayes, is about mining blockchains and describes how blockchains function. We use naive Bayes to optimize the blocks of supply chain management (SCM) blockchains by predicting transactions to anticipate storage levels.

Chapter 8, Solving the XOR Problem with a Feedforward Neural Network, is about building a feedforward neural network (FNN) from scratch to solve the XOR linear separability problem. The business case describes how to group orders for a factory.

Chapter 9, Abstract Image Classification with Convolutional Neural Networks (CNNs), describes CNN in detail: kernels, shapes, activation functions, pooling, flattening, and dense layers. The case study illustrates the use of a CNN using a webcam on a conveyor belt in a food-processing company.

Chapter 10, Conceptual Representation Learning, explains conceptual representation learning (CRL), an innovative way to solve production flows with a CNN transformed into a CRL metamodel (CRLMM). The case study shows how to use a CRLMM for transfer and domain learning, extending the model to other applications.

Chapter 11, Combining Reinforcement Learning and Deep Learning, combines a CNN with an MDP to build a solution for automatic planning and scheduling with an optimizer with a rule-based system.

The solution is applied to apparel manufacturing showing how to apply AI to real-life systems.

Chapter 12, AI and the Internet of Things (IoT), explores a support vector machine (SVM) assembled with a CNN. The case study shows how self-driving cars can find an available parking space automatically.

Chapter 13, Visualizing Networks with TensorFlow 2.x and TensorBoard, extracts information of each layer of a CNN and displays the intermediate steps taken by the neural network. The output of each layer contains images of the transformations applied.

Chapter 14, Preparing the Input of Chatbots with Restricted Boltzmann Machines (RBM) and Principal Component Analysis (PCA), explains how to produce valuable information using an RBM and a PCA to transform raw data into chatbot-input data.

Chapter 15, Setting Up a Cognitive NLP UI/CUI Chatbot, describes how to build a Google Dialogflow chatbot from scratch using the information provided by an RBM and a PCA algorithm. The chatbot will contain entities, intents, and meaningful responses.

Chapter 16, Improving the Emotional Intelligence Deficiencies of Chatbots, explains the limits of a chatbot when dealing with human emotions. The Emotion options of Dialogflow will be activated along with Small Talk to make the chatbot friendlier.

Chapter 17, Genetic Algorithms in Hybrid Neural Networks, enters our chromosomes, finds our genes, and helps you understand how our reproduction process works. From there, it is shown how to implement an evolutionary algorithm in Python, a genetic algorithm (GA). A hybrid neural network will show how to optimize a neural network with a GA.

Chapter 18, Neuromorphic Computing, describes what neuromorphic computing is and then explores Nengo, a unique neuromorphic framework with solid tutorials and documentation.

This neuromorphic overview will take you into the wonderful power of our brain structures to solve complex problems.

Chapter 19, Quantum Computing, will show quantum computers are superior to classical computers, what a quantum bit is, how to use it, and how to build quantum circuits. An introduction to quantum gates and example programs will bring you into the futuristic world of quantum mechanics.

Appendix, Answers to the Questions, provides answers to the questions listed in the Questions section in all the chapters.