Machine learning is a subfield of artificial intelligence (AI) concerned with the study and development of computer systems that learn from data and make predictions or decisions without explicit programming of the pathway between input and output. It involves training algorithms on appropriate datasets to recognize patterns and relationships within the data. In contrast with traditional AI, where output is a function of input and program, in machine learning the program is a function of the input and the output (at lest in the supervised case -- see below).

There are many ways to subdivide the field. A popular one is based on the availability of data labels:

1. Supervised Learning: This involves training models using labeled data to make predictions on unseen data. Applications include image recognition, song recognition, and spam email filtering.
2. Unsupervised Learning: In this case the algorithms explore unlabeled data to discover patterns or structures. Applications include customer segmentation and dimensionality reduction (reducing the "size" of each data point).

Machine learning is becoming an integral part of many technologies indispensable to modern life, such as search and recommendation, drug discovery, disease diagnosis, spam detection, voice transcription, text translation, and media generation.