Abstract
Artificial intelligence can be seen as the intelligence exhibited by machines. For an artificial intelligence system to be able to take decisions based on the data available, different type of learning methods, such as machine learning, need to be applied. Machine learning is a learning technique that gives machines the ability to learn without being explicitly programmed. It addresses the creation and study of algorithms that are capable of learning from data and making predictions about it. Machine learning algorithms can be divided into different categories including supervised learning, unsupervised learning, semisupervised learning, and reinforcement learning. In this article, the authors want to explain what machine learning is, as well as clearly establish the differences and relationship of machine learning with other related concepts, including artificial intelligence and deep learning. In addition, some possible use cases and applications will be named in order to provide the reader with a clear idea of what the potential of machine learning is.
TopMachine Learning
Machine Learning is a learning technique that gives machines the ability to learn without being explicitly programmed. It addresses the study and creation of algorithms that are capable of learning from data and making predictions about it. It is also important to notice that Machine Learning is seen as a subset of Artificial Intelligence, and not the other way around.
According to Raz, Llinas, Mittu and Lawless (2020), the basic idea behind Machine Learning methods is that a computer algorithm is trained to learn the behavior presented as part of previous experience and/or dataset to the extent that an outcome can be produced by the computer algorithm when it is presented with a never-before-seen dataset or situation.
Key Terms in this Chapter
Generative Adversarial Network: A type of deep neural network framework made of two neural networks (generator and discriminator) which compete against each other, and use a cooperative zero-sum game to learn. The generator is trained to produce fake data, and the discriminator tries to differentiate the generator’s fake data from real examples.
Artificial Neural Network: A computing system inspired by the biological neural networks that constitute a human brain.
Machine Learning: It refers to a learning technique that gives machines the ability to learn without being explicitly programmed. It is seen as a subset of Artificial Intelligence.
Convolutional Neural Network: A type of deep neural network designed for processing structured arrays of data and most commonly applied to analyze visual imagery because its ability to identify patterns in the input image, such as lines, gradients, circles, or even eyes and faces.
Deep Learning: It refers to Artificial Neural Networks and related Machine Learning algorithms that uses multiple layers of neurons. It is seen as a subset of Machine Learning in Artificial Intelligence.
Long Short-Term Memory: A type of deep neural network with a Recurrent Neural Network (RNN) architecture that, unlike standard feedforward neural networks, has feedback connections, and can process not only single data points (such as images), but also entire sequences of data (such as speech or video).
Unsupervised Learning: It is a type of Machine Learning. It uses learning algorithms to analyze and cluster unlabeled datasets. These algorithms focus on discovering hidden patterns or data groupings without the need for human intervention.
Reinforcement Learning: It is a type of Machine Learning. The algorithm discovers through its own experiences which actions produce the greatest rewards.
Supervised Learning: It is a type of Machine Learning. It is characterized by the use of labeled datasets to train algorithms that classify data or predict results accurately.
Semisupervised Learning: It is a type of Machine Learning. A hybrid of Supervised and Unsupervised learning techniques that combines labeled and unlabeled data during training.