Machine Learning

Machine learning algorithms have become an essential part of 5G communications and networks because they help us with resource allocation and scheduling, system monitoring, traffic management, fault prediction, network security, optimization, and others. Machine learning algorithms are composed of representation, evaluation, and optimization. This chapter illustrates the relationships of artificial intelligence (AI), machine learning and deep learning. Artificial intelligent systems imitate human behavior by using the models. Machine learning as one aspect of AI is a data analysis method for extracting knowledge from data. Deep learning as a subset of machine learning is inspired by the human brain structure, including interconnection of many neurons. The chapter discusses two categories of machine learning: supervised and unsupervised learning. Reinforcement learning (RL) deals with sequential decision‐making problems under uncertainty. RL uses training data including state, action, transition probability, reward, and discount, and develops an optimal policy to maximize cumulative rewards.