Distilled Machine Learning Algorithms in C++: A Primer on Essential Techniques

By Antonino Calderone

This book strives to make machine learning (ML) concepts more accessible to developers proficient in C++ who are eager to integrate these technologies into their projects. Utilizing the nunn library and TensorFlow, along with MATLAB (R), this guide aims to clarify key ML algorithms, focusing on a practical approach that links theoretical concepts with real-world applications. The content is tailored to be straightforward and effective, ensuring that complex ideas become approachable for those new to ML but skilled in C++. Book Structure Each chapter of this book is designed to not only educate but also to inspire practical experimentation and engagement with the material presented. This book is an invitation to actively participate in the unfolding landscape of machine learning, equipped with the knowledge and tools provided herein to apply these techniques effectively in various real-world scenarios. Here is an outline of the chapters and key topics covered in this book: Chapter 1: Introduction -- This chapter offers a comprehensive overview of the evolution of artificial intelligence, charting its key milestones from early concepts to modern advancements. It introduces fundamental aspects of machine learning such as data representation, learning algorithms, and their applications across various fields. Additionally, the concluding section of this chapter delves into the practical applications of TensorFlow, demonstrating its versatility and power in developing machine learning models with both C++ and Python. Chapter 2: The Artificial Neuron -- Explores the foundations of neural networks, starting with the concept of the biological neuron and extending to the Rosenblatt Perceptron and its limitations. Chapter 3: Foundations of Deep Learning with Multi-layer Neural Networks -- Introduces the architecture and functioning of multi-layer neural networks, including discussions on optimization and the backpropagation process. Chapter 4: Multi-layer Neural Networks Implementation in C++ -- Focuses on the practical aspects of coding multi-layer perceptrons in C++, with examples like solving the XOR problem and predicting Titanic survivorship. Chapter 5: Recognizing Handwritten Digits -- Builds upon the multi-layer perceptron (MLP) framework established in Chapter 4 to develop and implement a neural network tailored for recognizing handwritten digits. Chapter 6: Unsupervised Learning -- Covers key unsupervised learning techniques such as Hopfield networks and K-means clustering, emphasizing their theory and applications. Chapter 7: Exploring Complex Neural Network Architectures -- Discusses more complex neural network designs such as RNNs, LSTM networks, and Convolutional Neural Networks. Chapter 8: Reinforcement Learning -- Provides a deep dive into reinforcement learning methods like Q-Learning and SARSA, detailing their application in decision-making processes. Appendix A: Compendium of Additional Machine Learning Algorithms - An overview of additional essential algorithms that, although not treated in detail, form the cornerstone of many machine learning applications. Appendix B: Using MATLAB for Machine Learning -- An overview of machine learning techniques in MATLAB, offering guidance on algorithm selection and model deployment. Appendix C: Introduction to the nunn Library -- Provides a detailed look at the nunn library, its features, and its application in machine learning projects. Appendix D: nunn Library Helper Classes -- Describes helper classes within the nunn library.

• Format: Paperback
• Language: English
• ISBN: B0D3BDPH2H
• ISBN13: 9798322999638
• Release Date: May 2024
• Publisher: Independently published
• Length: 343 Pages
• Weight: 1.65 lbs.
• Dimensions: 10.0" x 0.8" x 7.0"