Energy Efficiency – from both economic and environmental points of view, it is important to minimize the energy costs of both training and running an agent or model.
Ensemble averaging in machine learning, particularly in the creation of artificial neural networks, is the process of creating multiple models and combining them to produce a desired output, as opposed to creating just one model[462 - Ensemble averaging [Электронный ресурс] www.engati.com URL: https://www.engati.com/glossary/ensemble-averaging (дата обращения 08.03.2022)].
Ensemble is a merger of the predictions of multiple models. You can create an ensemble via one or more of the following: different initializations; different hyperparameters; different overall structure. Deep and wide models are a kind of ensemble[463 - Ensemble [Электронный ресурс] https://www.primeclasses.in URL: https://www.primeclasses.in/glossary/data-science-course/machine-learning/ensemble (дата обращения: 27.03.2023)].
Enterprise Imaging has been defined as «a set of strategies, initiatives and workflows implemented across a health- care enterprise to consistently and optimally capture, index, manage, store, distribute, view, exchange, and analyze all clinical imaging and multimedia content to enhance the electronic health record» by members of the HIMSSSIIM Enterprise Imaging Workgroup[464 - Enterprise Imaging [Электронный ресурс] www.impact-advisors.com URL: https://www.impact-advisors.com/infrastructure/lessons-learned-while-implementing-a-vendor-neutral-archive-vna/ (дата обращения 22.02.2022)].
Entity annotation – the process of labeling unstructured sentences with information so that a machine can read them. This could involve labeling all people, organizations and locations in a document, for example[465 - Entity annotation [Электронный ресурс] https://bigdataanalyticsnews.com URL: https://bigdataanalyticsnews.com/artificial-intelligence-glossary/ (дата обращения: 27.03.2023)].
Entity extraction is an umbrella term referring to the process of adding structure to data so that a machine can read it. Entity extraction may be done by humans or by a machine learning model[466 - Entity extraction [Электронный ресурс] https://www.telusinternational.com URL: https://www.telusinternational.com/insights/ai-data/article/50-beginner-ai-terms-you-should-know (дата обращения: 09.04.2023)].
Entropy — the average amount of information conveyed by a stochastic source of data[467 - Entropy [Электронный ресурс] https://appen.com URL: https://appen.com/ai-glossary/ (дата обращения 28.02.2022)].
Environment in reinforcement learning, the world that contains the agent and allows the agent to observe that world’s state. For example, the represented world can be a game like chess, or a physical world like a maze. When the agent applies an action to the environment, then the environment transitions between states[468 - Environment [Электронный ресурс] https://developers.google.com URL: https://developers.google.com/machine-learning/glossary#environment (дата обращения: 16.06.2023)].
Episode in reinforcement learning, is each of the repeated attempts by the agent to learn an environment[469 - Episode [Электронный ресурс] https://developers.google.com URL: https://developers.google.com/machine-learning/glossary#episode (дата обращения: 11.07.2023)].
Epoch in the context of training Deep Learning models, is one pass of the full training data set[470 - Эпоха (Epoch) [Электронный ресурс] https://tgdratings.com URL: https://tgdratings.com/ru/glossary/epoch/ (дата обращения: 11.07.2023)],[471 - Epoch [Электронный ресурс] https://developers.google.com URL: https://developers.google.com/machine-learning/glossary#epoch (дата обращения: 11.07.2023)].
Epsilon greedy policy in reinforcement learning, is a policy that either follows a random policy with epsilon probability or a greedy policy otherwise. For example, if epsilon is 0.9, then the policy follows a random policy 90% of the time and a greedy policy 10% of the time[472 - Epsilon greedy policy [Электронный ресурс] https://developers.google.com URL: https://developers.google.com/machine-learning/glossary#epsilon-greedy-policy (дата обращения: 11.07.2023)].
Equality of opportunity is a fairness metric that checks whether, for a preferred label (one that confers an advantage or benefit to a person) and a given attribute, a classifier predicts that preferred label equally well for all values of that attribute. In other words, equality of opportunity measures whether the people who should qualify for an opportunity are equally likely to do so regardless of their group membership. For example, suppose Glubbdubdrib University admits both Lilliputians and Brobdingnagians to a rigorous mathematics program. Lilliputians’ secondary schools offer a robust curriculum of math classes, and the vast majority of students are qualified for the university program. Brobdingnagians’ secondary schools don’t offer math classes at all, and as a result, far fewer of their students are qualified. Equality of opportunity is satisfied for the preferred label of «admitted» with respect to nationality (Lilliputian or Brobdingnagian) if qualified students are equally likely to be admitted irrespective of whether they’re a Lilliputian or a Brobdingnagian[473 - Equality of opportunity [Электронный ресурс] https://developers.google.com URL: https://developers.google.com/machine-learning/glossary#equality-of-opportunity (дата обращения: 29.06.2023)].
Equalized odds is a fairness metric that checks if, for any particular label and attribute, a classifier predicts that label equally well for all values of that attribute[474 - Equalized odds [Электронный ресурс] https://developers.google.com URL: https://developers.google.com/machine-learning/glossary#equalized-odds (дата обращения 04.07.2023)].
Ergatic system is a scheme of production, one of the elements of which is a person or a group of people and a technical device through which a person carries out his activities. The main features of such systems are socio-psychological aspects. Along with the disadvantages (the presence of the «human factor»), ergatic systems have a number of advantages, such as fuzzy logic, evolution, decision-making in non-standard situations[475 - Эргатическая система [Электронный ресурс] https://en.wikipedia.org URL: https://ru.wikipedia.org/wiki/Эргатическая_система (дата обращения: 07.07.2022)].
Error backpropagation – the process of adjusting the weights in a neural network by minimizing the error at the output. It involves a large number of iteration cycles with the training data[476 - Error backpropagation [Электронный ресурс] https://neurohive.io URL: https://neurohive.io/ru/osnovy-data-science/obratnoe-rasprostranenie/ (дата обращения: 31.01.2022)].
Error-driven learning is a sub-area of machine learning concerned with how an agent ought to take actions in an environment so as to minimize some error feedback. It is a type of reinforcement learning[477 - Error-driven learning [Электронный ресурс] https://en.wikipedia.org URL: https://en.wikipedia.org/wiki/Error-driven_learning (дата обращения: 16.06.2023)].
Ethical use of artificial intelligence is a systematic normative understanding of the ethical aspects of AI based on an evolving complex, comprehensive and multicultural system of interrelated values, principles and procedures that can guide societies in matters of responsible consideration of the known and unknown consequences of the use of AI technologies for people, communities, the natural environment environment and ecosystems, as well as serve as a basis for decision-making regarding the use or non-use of AI-based technologies[478 - Доклад комиссии по социальным и гуманитарным наукам (SHS). [Электронный ресурс] https://unesdoc.unesco.org URL: https://unesdoc.unesco.org/ark:/48223/pf0000379920_rus.page=16 (дата обращения: 29.01.2022)].
Ethics of Artificial Intelligence is the ethics of technology specific to robots and other artificial intelligence beings, which is divided into robot ethics and machine ethics. The former one is about the concern with the moral behavior of humans as they design, construct, use, and treat artificially intelligent beings, and the latter one is about the moral behavior of artificial moral agents[479 - Этика искусственного интеллекта [Электронный ресурс] https://ru.wikipedia.org URL: https://ru.wikipedia.org/wiki/Этика_искусственного_интеллекта (дата обращения: 11.07.2023)].
Evolutionary algorithm (EA) is a subset of evolutionary computation, a generic population-based metaheuristic optimization algorithm. An EA uses mechanisms inspired by biological evolution, such as reproduction, mutation, recombination, and selection. Candidate solutions to the optimization problem play the role of individuals in a population, and the fitness function determines the quality of the solutions (see also loss function). Evolution of the population then takes place after the repeated application of the above operators[480 - Evolutionary algorithm [Электронный ресурс] https://wiki.loginom.ru URL: https://wiki.loginom.ru/articles/evolution-algorithm.html (дата обращения: 08.02.2022)].
Evolutionary computation is a family of algorithms for global optimization inspired by biological evolution, and the subfield of artificial intelligence and soft computing studying these algorithms. In technical terms, they are a family of population-based trial and error problem solvers with a metaheuristic or stochastic optimization character[481 - Evolutionary computation [Электронный ресурс] https://en.wikipedia.org URL: https://en.wikipedia.org/wiki/Evolutionary_computation (дата обращения: 10.07.2023)].
Evolving classification function (ECF) – evolving classifier functions or evolving classifiers are used for classifying and clustering in the field of machine learning and artificial intelligence, typically employed for data stream mining tasks in dynamic and changing environments[482 - Evolving classification function [Электронный ресурс] www.semanticscholar.org URL: https://www.semanticscholar.org/topic/Evolving-classification-function/344460 (дата обращения 28.02.2022)].
Example – one row of a dataset. An example contains one or more features and possibly a label. See also labeled example and unlabeled example[483 - Example [Электронный ресурс] https://www.primeclasses.in URL: https://www.primeclasses.in/glossary/data-science-course/machine-learning/example (дата обращения: 28.06.2023)].
Executable – executable code, an executable file, or an executable program, sometimes simply referred to as an executable or binary, causes a computer «to perform indicated tasks according to encoded instructions», as opposed to a data file that must be interpreted (parsed) by a program to be meaningful[484 - Executable [Электронный ресурс] https://en.wikipedia.org URL: https://en.wikipedia.org/wiki/Executable (дата обращения: 07.07.2022)].
Existential risk – the hypothesis that substantial progress in artificial general intelligence (AGI) could someday result in human extinction or some other unrecoverable global catastrophe[485 - Existential risk [Электронный ресурс] https://en.wikipedia.org URL: https://en.wikipedia.org/wiki/Existential_risk_from_artificial_general_intelligence (дата обращения: 10.07.2023)].
Experience replay in reinforcement learning, a DQN technique used to reduce temporal correlations in training data. The agent stores state transitions in a replay buffer, and then samples transitions from the replay buffer to create training data[486 - Experience replay [Электронный ресурс] https://developers.google.com URL: https://developers.google.com/machine-learning/glossary#experience-replay (дата обращения: 16.06.2023)].
Experimenter’s bias it is the tester’s tendency to seek and interpret information, or give preference to one or another information, that is consistent with his point of view, belief or hypothesis. A kind of cognitive distortion and bias in inductive thinking[487 - Experimenter’s bias [Электронный ресурс] https://developers.google.com URL: https://developers.google.com/machine-learning/glossary#confirmation-bias (дата обращения: 28.06.2023)].
Expert system is a computer system that emulates the decision-making ability of a human expert. Expert systems are designed to solve complex problems by reasoning through bodies of knowledge, represented mainly as if—then rules rather than through conventional procedural code[488 - Expert system [Электронный ресурс] https://en.wikipedia.org URL: https://en.wikipedia.org/wiki/Expert_system (дата обращения: 11.07.2023)],[489 - Экспертная система [Электронный ресурс] https://ru.wikipedia.org URL: https://ru.wikipedia.org/wiki/Экспертная_система (дата обращения: 11.07.2023)].
Expert systems are systems that use industry knowledge (from medicine, chemistry, law) combined with sets of rules that describe how to apply the knowledge[490 - Экспертные системы [Электронный ресурс] https://apr.moscow URL: https://apr.moscow/content/data/6/11 Технологии искусственного интеллекта. pdf стр. 7 (дата обращения: 11.07.2023)].
Explainable artificial intelligence (XAI) is a key term in AI design and in the tech community as a whole. It refers to efforts to make sure that artificial intelligence programs are transparent in their purposes and how they work. Explainable AI is a common goal and objective for engineers and others trying to move forward with artificial intelligence progress[491 - Объяснимый искусственный интеллект [Электронный ресурс] https://cyberleninka.ru URL: https://cyberleninka.ru/article/n/sovremennyy-etap-razvitiya-iskusstvennogo-intellekta-ii-i-primenenie-metodov-i-sistem-ii-v-energetike (дата обращения: 16.06.2023)]
Вы ознакомились с фрагментом книги.
Для бесплатного чтения открыта только часть текста.
Приобретайте полный текст книги у нашего партнера:
