Author: Zen and the Art of Computer Programming
1 Introduction
##1.1 What is machine learning?
It refers to a series of automated machine learning algorithms that use computer programming, statistical methods, or probability theory to predict, analyze, and train data, and finally achieve knowledge acquisition from data, improvement of behavior, or prediction of event development results. The main purpose of machine learning is to enable computers to have "learning ability", so as to better complete tasks, solve problems, improve efficiency, and reduce costs. In application scenarios such as image recognition, text classification, spam filtering, medical diagnosis, and automatic driving play a pivotal role in other fields.
1.1.1 Definition of Machine Learning
Machine learning (Machine Learning) is a multi-field interdisciplinary subject, covering artificial intelligence, cognitive science, computer science, pattern recognition, computing theory, communication engineering and other fields. It is a technique that allows computer systems to learn and adapt to their environment. Machine learning includes three subfields: supervised learning, unsupervised learning, and reinforcement learning.
1.1.2 Three elements of machine learning
- Data: The data set required by a machine learning algorithm usually consists of input variables (features) and output variables.
- Model: A machine learning model is a function that maps input variables to output variables.
- Algorithm: The process of deciding how to learn and build a model from a given input data is an algorithm. Different algorithms will affect the accuracy, efficiency, and generalization ability of the learning results.
1.1.3 Machine Learning Classification
According to the different learning methods, machine learning can be divided into three categories:
- Supervised learning: By giving input-output sample data, a model is trained using existing label information, so that the model can predict unknown data. Supervised learning can be divided into supervised learning and semi-supervised learning.
- Supervised learning (Semi-supervised l