Machine Learning

Updated Sept 14, 2020 · 4 min read

What is machine learning?

Machine learning (ML) is a technique used in artificial intelligence where engineers train algorithms to learn patterns in large amounts of data. Based on what they learn, algorithms can perform tasks, such as marking emails as spam (email filtering) or recognizing letters and numbers when checks are scanned at ATMs (computer vision).

Unlike most conventional forms of development, machine learning doesn’t rely on engineers explicitly programming rules and behaviors into the algorithms. Instead, the algorithms use mathematical and statistical techniques to learn behaviors solely from data.

Approaches to machine learning

Machine learning can broadly be broken down into the following paradigms:

  • Supervised learning: An algorithm is presented with examples of inputs and desired outputs and then learns rules for connecting the inputs to outputs.
  • Unsupervised learning: An algorithm isn’t presented with examples. Instead, it learns rules by finding structure and commonalities in data.
  • Reinforcement learning (RL): An algorithm learns by trying to perform some goal and is rewarded for the behavior it ought to perform. The algorithm tries to maximize rewards.

Related Terms

Data Science
Data Science

Data science is an area of study that focuses on organizing, analyzing, and extracting insights from large data sets (big data) using...

Data Science
Data Store

A data store (or datastore) is a central location for storing and managing sets of data, such as in a database or file system.