The StartOnAI Book

Summary

A Guide to Machine Learning, Deep Learning, and their Applications acts as a comprehensive guide to the innovative fields of machine learning and deep learning. The tutorial-style prose allows the reader to explore the topics from scratch and to broaden their knowledge of the theory and for practical understanding. By the time you are finished with this book, you will be confident with your knowledge of what machine learning is and how state-of-the-art algorithms can be utilized to solve real-world problems. Whether you are a student (high school, undergraduate, or graduate) or a data scientist, this book will supplement your current knowledge with rigorous yet practical tools to break down machine learning and familiarize yourself with best practices used in the industry. 

About Machine Learning

In today’s world, machine learning is becoming increasingly ubiquitous as the forefront of methods for pattern recognition. These techniques require input data and allow the computer to learn from experience. Through the recent advances in data science software and the availability of libraries, it is now possible for programmers and basic users to get familiar with creating their own programs that are capable of learning from the data. Deep learning is the fastest-growing subset of machine learning and is loosely inspired by biological networks. 

Why this book?

This book serves as a guide for both beginners and users of machine learning to become more familiar with the fields on both a theoretical basis and through a hands-on, practical example-based approach. Unlike other books on the subject, we use efficient, grounded tools to demonstrate the power of machine learning while guiding the reader from the fundamentals of data science and mathematics towards state-of-the-art algorithms and modern developments in the field. What’s special about this book is that the applications are at the center of the content and detailed commentary is present in every chapter. We also prioritize high-level, concrete explanations over less intuitive pathways to understanding. Moreover, this book was written by students and is for students.

What’s Inside?

• Introduction to Linear Algebra for Machine Learning, Python, Calculus, Probability, and Data Structures 
• Supervised Learning fundamentals with linear regression, gradient descent, logistic regression, decision trees, random forests, k-nearest neighbors, perceptron, and support vector machines
• Advanced Classification techniques with Bayesian machine learning, dimensionality reduction, kernel functions, linear discriminant analysis, and principal component analysis
• Deep Learning basics with multi-layer perceptron, forward propagation, backpropagation, activation functions, and artificial neural networks
• Advanced Deep Learning with L2 and dropout regularization, batch normalization, vanishing gradients, mini-batch gradient descent, recurrent neural networks, LSTMs, and generative adversarial networks
• Practical tools to deconstruct deep learning with coding examples in NumPy, SciPy, Keras, and TensorFlow
• Direct applications to reinforcement learning, computer vision, image classification, sound classification, natural language processing, genomics, and industrial uses

For the Reader

With just knowledge of high school mathematics, readers can follow along. No knowledge of calculus, linear algebra, or programming is required, just a willingness to learn. This book builds understanding through intuitive explanations and direct applications. Through this content, you will learn both challenging yet elegant concepts to practice the application of machine learning to computer vision, natural language processing, and bioinformatics.