Intro
Intro¶
Welcome to the world of machine learning. It is a complex world with still much to be discovered. This book will attempt to simplify its many complex concepts into words that are easy to comprehend and visualize.
Machine learning is about trying to find patterns in data, and using those patterns to make predictions and generalize to unseen data. For example, the way you learned to add and subtract is by seeing examples, learning the patterns behind how to add single-digit numbers, and being able to generalize this to be able to solve any addition/subtraction problem. We’ll find that the machine learning algorithms we learn behave in a very similar way.
How can we find these patterns in data? We assume certain models (basically, the algorithm) for our given data, then fit those models to the data to gain a better understanding of these patterns. Optomization algorithms essentially “learn” the most likely pattern that from the data.
The most important part of all, though, is the data itself. For quality machine learning, we not only need large amounts of data, but we also need quality data. A term you may have heard thrown around is Big Data: this just means that there’s an absolutely massive amount of data available for many different applications. We want to make sense of this data.
From Netflix recommendations to self-driving cars, machine learning is at the heart of almost all of the modern scientific advances we know. Let’s dive into it.