– Federal Law of July 27, 2006 №152 (as amended on April 24, 2020) “On Personal Data” [[10 - Федеральный закон от 27.07.2006 N 152-ФЗ (ред. от 24.04.2020) “О персональных данных”. [Электронный ресурс] // legalacts.ru URL: https://legalacts.ru/doc/152_FZ-o-personalnyh-dannyh/ (https://legalacts.ru/doc/152_FZ-o-personalnyh-dannyh/)]].
– National program “Digital Economy of the Russian Federation” [[11 - Национальная программа “Цифровая экономика Российской Федерации”. Министерство цифрового развития, связи и массовых коммуникаций Российской Федерации. [Электронный ресурс] // digital.gov.ru. URL: https://digital.gov.ru/ru/activity/directions/858/ (https://digital.gov.ru/ru/activity/directions/858/)]].
– State Program “Digital Economy of the Russian Federation” [[12 - Государственная Программа “Цифровая экономика Российской Федерации”. [Электронный ресурс] // static.government.ru URL: http://static.government.ru/media/files/9gFM4FHj4PsB79I5v7yLVuPgu4bvR7M0.pdf (http://static.government.ru/media/files/9gFM4FHj4PsB79I5v7yLVuPgu4bvR7M0.pdf)]].
1000 terms and definitions.
Is it a lot or a little?
Our experience suggests that for mutual understanding it is enough for two interlocutors to know a dozen or a maximum of two dozen definitions, but when it comes to professional activities, it may turn out that it is not enough to know even a few dozen terms.
This book contains the terms, in our opinion, the most frequently used, both in everyday work and professional activities by specialists of various professions interested in the topic of “artificial intelligence”.
In conclusion, I would like to add and inform the dear reader that we have tried very hard to make for you the necessary and useful “product” and “tool”.
35th Moscow International Book Fair
The first version of the book was presented by us at the 35th Moscow International Book Fair in 2022.
This book is a completely open and free document for distribution. If you use it in your practical work, please make a link to this book.
Many of the terms and definitions for them in this book are found on the Internet. They are repeated dozens or hundreds of times on various information resources (mainly foreign ones). Nevertheless, we set ourselves the goal of collecting and systematizing the most relevant of them in one place from a variety of sources, translating and adapting the necessary ones into Russian, and rewriting some of them based on our own experience. In view of the foregoing, we do not claim authorship or uniqueness of the terms and definitions presented.
Links to primary sources are affixed to the original terms and definitions (that is, if the definition was originally in English, then the link is indicated after this definition). If the definition was given in Russian, translated into English and adapted, then the reference is not indicated (in this edition of the book). This book was written by Russian authors and therefore the translation of terms into Russian is given in brackets.
We continue to work on improving the quality and content of the text of this book, including supplementing it with new knowledge in the subject area. We will be grateful for any feedback, suggestions and clarifications. Please send them to aleksander.chesalov@yandex.ru
Happy reading and productive work!
Yours, Alexander Chesalov, Alexander Vlaskin and Matvey Bakanach.
09/22/2022
ARTIFICIAL INTELLIGENCE GLOSSARY
“A”
A/B Testing (A/B-тестирование) – A statistical way of comparing two (or more) techniques, typically an incumbent against a new rival. A/B testing aims to determine not only which technique performs better but also to understand whether the difference is statistically significant. A/B testing usually considers only two techniques using one measurement, but it can be applied to any finite number of techniques and measures [[13 - A/B Testing [Electronic resource] // vwo.com URL: https://vwo.com/ab-testing/ (date of the application: 28.01.2022)]].
Abductive logic programming (ALP) (Абдуктивное логическое программирование) – A high-level knowledge-representation framework that can be used to solve problems declaratively based on abductive reasoning. It extends normal logic programming by allowing some predicates to be incompletely defined, declared as adducible predicates [[14 - Abductive Logic Programming (ALP) [Electronic resource] // engati.com URL https://www.engati.com/glossary/abductive-logic-programming (https://www.engati.com/glossary/abductive-logic-programming) (date of the application 14.02.2022)]].
Abductive reasoning (Also abduction) (Абдукция) — A form of logical inference which starts with an observation or set of observations then seeks to find the simplest and most likely explanation. This process, unlike deductive reasoning, yields a plausible conclusion but does not positively verify it. abductive inference, or retroduction [[15 - Abductive reasoning [Electronic resource] // MRS BLOG URL: http://msrblog.com/science/mathematic/about-abductive-reasoning.html (http://msrblog.com/science/mathematic/about-abductive-reasoning.html) (date of the application 14.02.2022)]].
Abstract data type (Абстрактный тип данных) — A mathematical model for data types, where a data type is defined by its behavior (semantics) from the point of view of a user of the data, specifically in terms of possible values, possible operations on data of this type, and the behavior of these operations [[16 - Abstract data type [Electronic resource] // EMBEDDED ARTISTRY URL: https://embeddedartistry.com/fieldmanual-terms/abstract-data-type/ (date of the application 14.02.2022)]].
Abstraction (Абстракция) — The process of removing physical, spatial, or temporal details or attributes in the study of objects or systems in order to more closely attend to other details of interest.
Accelerating change (Ускорение изменений) — A perceived increase in the rate of technological change throughout history, which may suggest faster and more profound change in the future and may or may not be accompanied by equally profound social and cultural change [[17 - Accelerating change [Электронный ресурс] // ru.knowledgr.com (дата обращения: 14.02.2022)]].
Access to information (Доступ к информации) – the ability to obtain information and use it.
Access to information constituting a commercial secret (Доступ к информации, составляющей коммерческую тайну) – familiarization of certain persons with information constituting a commercial secret, with the consent of its owner or on other legal grounds, provided that this information is kept confidential.
Accuracy (Точность) – The fraction of predictions that a classification model got right.
Action (Действие) – In reinforcement learning, the mechanism by which the agent transitions between states of the environment. The agent chooses the action by using a policy.
Action language (Язык действий) — A language for specifying state transition systems, and is commonly used to create formal models of the effects of actions on the world. Action languages are commonly used in the artificial intelligence and robotics domains, where they describe how actions affect the states of systems over time, and may be used for automated planning [[18 - https://www.semanticscholar.org/topic/Action-language/72365 (https://www.semanticscholar.org/topic/Action-language/72365)]].
Action model learning (Обучение модели действий) – An area of machine learning concerned with creation and modification of software agent’s knowledge about effects and preconditions of the actions that can be executed within its environment. This knowledge is usually represented in logic-based action description language and used as the input for automated planners [[19 - Action model learning [Электронный ресурс] // Semantic Scholar URL: https://www.semanticscholar.org/topic/Action-model-learning/1677625 (дата обращения 14.02.2022)]].
Action selection (Выбор действия) — A way of characterizing the most basic problem of intelligent systems: what to do next. In artificial intelligence and computational cognitive science, “the action selection problem” is typically associated with intelligent agents and animats – artificial systems that exhibit complex behaviour in an agent environment [[20 - Action selection [Электронный ресурс] // https://www.netinbag.com/ URL: https://www.netinbag.com/ru/internet/what-is-action-selection.html (https://www.netinbag.com/ru/internet/what-is-action-selection.html) (дата обращения: 18.02.2022)]].
Activation function (Функция активации нейрона) – In the context of Artificial Neural Networks, a function that takes in the weighted sum of all of the inputs from the previous layer and generates an output value to ignite the next layer [[21 - https://appen.com/ai-glossary/ (https://appen.com/ai-glossary/)]].
Active Learning/Active Learning Strategy (Активное обучение/ Стратегия активного обучения) – is a special case of Semi-Supervised Machine Learning in which a learning agent is able to interactively query an oracle (usually, a human annotator) to obtain labels at new data points. A training approach in which the algorithm chooses some of the data it learns from. Active learning is particularly valuable when labeled examples are scarce or expensive to obtain. Instead of blindly seeking a diverse range of labeled examples, an active learning algorithm selectively seeks the particular range of examples it needs for learning.
Adam optimization algorithm (Алгоритм оптимизации Адам) – it is an extension of stochastic gradient descent which has recently gained wide acceptance for deep learning applications in computer vision and natural language processing [[22 - Adam optimization algorithm [Электронный ресурс] // archive.org URL: https://archive.org/details/riseofexpertcomp00feig (дата обращения: 11.03.2022)]].
Adaptive algorithm (Адаптивный алгоритм) – An algorithm that changes its behavior at the time it is run, based on a priori defined reward mechanism or criterion [[23 - Adaptive algorithm. [Электронный ресурс] // dic.academic.ru (дата обращения: 27.01.2022)]].
Adaptive Gradient Algorithm (AdaGrad) (Адаптивный градиентный алгоритм) – A sophisticated gradient descent algorithm that rescales the gradients of each parameter, effectively giving each parameter an independent learning rate [[24 - Adaptive Gradient Algorithm. [Электронный ресурс] // jmlr.org. URL: https://www.jmlr.org/papers/volume12/duchi11a/duchi11a.pdf (дата обращения: 18.02.2022)]].
Adaptive neuro fuzzy inference system (ANFIS) (Also adaptive network-based fuzzy inference system.) (Адаптивная система нейро-нечеткого вывода) – A kind of artificial neural network that is based on Takagi – Sugeno fuzzy inference system. The technique was developed in the early 1990s. Since it integrates both neural networks and fuzzy logic principles, it has potential to capture the benefits of both in a single framework. Its inference system corresponds to a set of fuzzy IF – THEN rules that have learning capability to approximate nonlinear functions. Hence, ANFIS is considered to be a universal estimator. For using the ANFIS in a more efficient and optimal way, one can use the best parameters obtained by genetic algorithm [[25 - Adaptive neuro fuzzy inference system (ANFIS) [Электронный ресурс] // hrpub.ru URL: https://www.hrpub.org/download/20190930/AEP1-18113213.pdf (дата обращения 14.02.2022)]].
Adaptive system (Адаптивная система) is a system that automatically changes the data of its functioning algorithm and (sometimes) its structure in order to maintain or achieve an optimal state when external conditions change.
Additive technologies (Аддитивные технологии) are technologies for the layer-by-layer creation of three-dimensional objects based on their digital models (“twins”), which make it possible to manufacture products of complex geometric shapes and profiles.
Admissible heuristic (Допустимая эвристика) – In computer science, specifically in algorithms related to pathfinding, a heuristic function is said to be admissible if it never overestimates the cost of reaching the goal, i.e., the cost it estimates to reach the goal is not higher than the lowest possible cost from the current point in the path.
Affective computing (Also artificial emotional intelligence or emotion AI.) (Аффективные вычисления) – The study and development of systems and devices that can recognize, interpret, process, and simulate human affects. Affective computing is an interdisciplinary field spanning computer science, psychology, and cognitive science [[26 - Affective computing [Электронный ресурс] // OpenMind URL: https://www.bbvaopenmind.com/en/technology/digital-world/what-is-affective-computing/ (дата обращения 14.02.2022)]].
Agent (Агент) – In reinforcement learning, the entity that uses a policy to maximize expected return gained from transitioning between states of the environment.
Agent architecture (Архитектура агента) – A blueprint for software agents and intelligent control systems, depicting the arrangement of components. The architectures implemented by intelligent agents are referred to as cognitive architectures [[27 - Agent architecture [Электронный ресурс] // dic.academic URL: https://en-academic.com/dic.nsf/enwiki/2205509 (дата обращения 28.02.2022)]].
Agglomerative clustering (See hierarchical clustering.) (Агломеративная кластеризация) – Agglomerative clustering first assigns every example to its own cluster, and iteratively merges the closest clusters to create a hierarchical tree.
Aggregate (Агрегат) A total created from smaller units. For instance, the population of a county is an aggregate of the populations of the cities, rural areas, etc., that comprise the county. To total data from smaller units into a large unit. [[28 - Aggregate [Электронный ресурс] www.umich.edu URL: https://www.icpsr.umich.edu/web/ICPSR/cms/2042#A (https://www.icpsr.umich.edu/web/ICPSR/cms/2042#A) (дата обращения: 07.07.2022)]]
Aggregator (Агрегатор) A feed aggregator is a type of software that brings together various types of Web content and provides it in an easily accessible list. Feed aggregators collect things like online articles from newspapers or digital publications, blog postings, videos, podcasts, etc. A feed aggregator is also known as a news aggregator, feed reader, content aggregator or an RSS reader. [[29 - Aggregator [Электронный ресурс] www.techopedia.com URL: https://www.techopedia.com/definition/2502/feed-aggregator (https://www.techopedia.com/definition/2502/feed-aggregator) (дата обращения: 07.07.2022)]]
AI benchmark (Исходная отметка (Бенчмарк) ИИ) is an AI benchmark for evaluating the capabilities, efficiency, performance and for comparing ANNs, machine learning (ML) models, architectures and algorithms when solving various AI problems, special benchmarks are created and standardized, initial marks. For example, Benchmarking Graph Neural Networks – benchmarking (benchmarking) of graph neural networks (GNS, GNN) – usually includes installing a specific benchmark, loading initial datasets, testing ANNs, adding a new dataset and repeating iterations.
AI chipset market (Рынок чипсетов ИИ) is the market for chipsets for artificial intelligence (AI) systems.
AI acceleration (ИИ ускорение) – acceleration of calculations encountered with AI, specialized AI hardware accelerators are allocated for this purpose (see also artificial intelligence accelerator, hardware acceleration).
