Demystifying Machine Learning | Glossary

Simple glossary of terms, abbreviations, and concepts related to the field and sub-fields of artificial intelligence. Based on technical definitions by pioneers and leaders in AI.

Relevant fields: [ AI ] [ machine learning ] [ deep learning] [ data science ] [ computer vision ] [ NLP ] [ human AI interaction ] [ robotics ]

Below please find a searchable glossary of terms that are relevant to the scientific field and sub-fields of artificial intelligence — on ongoing growth.

This glossary is by no means complete, if a term it’s missing, please let me know in the comments and I’ll add its definition to this growing glossary. Also, if you have feedback regarding a definition of any of the term(s), please let me know as well.

Glossary of Terms | Sorted (A-Z)

Agent: A bot used in AI related tasks.

Algorithm: Process that follows a set of rules, a problem solver — especially used by computers.

AlphaGo: Computer program that plays the board game Go. Recognized as the first computer Go program to beat a professional human Go player [5].

Artificial Intelligence (AI): Science and engineering of making computers behave in ways that, until recently, we thought required human intelligence [1].

Autonomous: Device or tool capable of operating without direct human control.

Backpropagation: A way in which neural networks train, during training they find input and provide an output, then adjust the

Black box: Complex neural network in which algorithms, contents and decision-making processes are unknown to the end-user.

Bot: An autonomous program that can interact with computer systems, programs or users. Mostly, directly or indirectly supervised by a human.

Clustering: Unsupervised machine learning.

Computational learning theory: Sub-field of AI devoted to studying the design and analysis of machine learning algorithms [4].

Computer program: Collection of instructions that perform specific tasks when launched by a computer.

Computer science: Scientific study of the principles and use of computers.

Computer vision: Interdisciplinary scientific sub-field of AI and CS that aims at giving computers a visual understanding of its input.

Convolutional neural network (CNN): Class of deep neural network used on image recognition, processing, and analysis — specifically designed to process pixel data.

Data: Digital collection of information.

Data mining: Process of examination and discovery of patterns in data to generate new information.

Datasets: A collection of related sets of data that is composed of separate elements but can be manipulated as a unit by a computer.

Data science: Interdisciplinary scientific field on processes and systems to extract knowledge or insights from data in its various forms.

Deep learning: Subset of machine learning in which layered neural networks, combined with high computing power and large datasets can create powerful machine learning models.

Deep neural network (DNN): Large computer system modeled after the human brain.

Dimensional reduction: Process of reducing the number of random variables by obtaining a set of principal variables via feature selection and/or feature extraction.

Explainable AI: AI capable to explain its decision making process.

Generative adversarial networks (GANs): Two neural networks contesting with each other in a zero-sum game framework.

Heuristics: Rules drawn from experience used to solve a problem faster than other traditional problem-solving methods in AI. While faster, a heuristic approach typically is less optimal than the classic methods it replaces.

Input: What is put in, taken in, or operated on a process or system.

Intelligence: Computational part of the ability to achieve goals in the world. Varying kinds and degrees of intelligence occur in people, many animals, and some machines.

Linear Algebra: Branch of mathematics concerning vector spaces and linear mappings between such spaces. It includes the study of lines, planes, and sub-spaces, yet is also concerned with properties common to all vector spaces.

Long short-term memory networks (LSTMs): Special kind of recurrent neural networks capable of learning long-term dependencies [2].

Machine learning (ML): Scientific branch of AI, which focuses on the study of computer algorithms that allow computer programs to automatically improve through experience [3].

Machine learning model/system: Question/answering system that takes care of processing machine learning related tasks.

Machine perception: Capability for a computer system to interpret data in a manner similar to the way humans use their senses to interact and relate to the world around them.

Natural Language Processing (NLP): Scientific branch of AI, which focuses in helping computers understand, interpret and manipulate human language.

Neural network: Computer system modeled after the human brain.

Recurrent neural network (RNN): Powerful and robust type of neural networks, capable of complex cycles in internal memory.

Turing test: Test that only passes if a human being is unable to distinguish a machine from a human.

DISCLAIMER: The views expressed in this article are those of the author(s) and do not represent the views of Carnegie Mellon University, nor other companies (directly or indirectly) associated with the author(s). These writings are not intended to be final products, yet rather a reflection of current thinking, along being a catalyst for discussion and improvement.


[1] Carnegie Mellon University Dean of Computer Science on the Future of AI | Forbes |

[2] Understanding LSTM Networks | Christopher Olah |

[3] Machine Learning Definition | Tom M. Mitchell| McGraw-Hill Science/Engineering/Math; (March 1, 1997), Page 1 |

[4] Computational Learning Theory | Wikipedia |

[5] AlphaGo: Mastering the ancient game of Go with machine learning | Google AI Blog |

Source: Artificial Intelligence on Medium