Глоссариум по искусственному интеллекту: 2500 терминов. Том 2. Александр Юрьевич Чесалов

Чтение книги онлайн.

Читать онлайн книгу Глоссариум по искусственному интеллекту: 2500 терминов. Том 2 - Александр Юрьевич Чесалов страница 12

Глоссариум по искусственному интеллекту: 2500 терминов. Том 2 - Александр Юрьевич Чесалов

Скачать книгу

rel="nofollow" href="#note161" type="note">161.

      Bigram – an N-grams in which N=2162.

      Binary choice regression model is a regression model in which the dependent variable is dichotomous or binary. Dependent variable can take only two values and mean, for example, belonging to a particular group163.

      Binary classification is a type of classification task that outputs one of two mutually exclusive classes. For example, a machine learning model that evaluates email messages and outputs either «spam» or «not spam» is a binary classifier164.

      Binary format is any file format in which information is encoded in some format other than a standard character-encoding scheme. A file written in binary format contains information that is not displayable as characters. Software capable of understanding the particular binary format method of encoding information must be used to interpret the information in a binary-formatted file. Binary formats are often used to store more information in less space than possible in a character format file. They can also be searched and analyzed more quickly by appropriate software. A file written in binary format could store the number «7» as a binary number (instead of as a character) in as little as 3 bits (i.e., 111), but would more typically use 4 bits (i.e., 0111). Binary formats are not normally portable, however. Software program files are written in binary format. Examples of numeric data files distributed in binary format include the IBM-binary versions of the Center for Research in Security Prices files and the U.S. Department of Commerce’s National Trade Data Bank on CD-ROM. The International Monetary Fund distributes International Financial Statistics in a mixed-character format and binary (packed-decimal) format. SAS and SPSS store their system files in binary format165.

      Binary number is a number written using binary notation which only uses zeros and ones. Example: Decimal number 7 in binary notation is: 111166.

      Binary tree is a tree data structure in which each node has at most two children, which are referred to as the left child and the right child. A recursive definition using just set theory notions is that a (non-empty) binary tree is a tuple (L, S, R), where L and R are binary trees or the empty set and S is a singleton set. Some authors allow the binary tree to be the empty set as well167.

      Binning is the process of combining charge from neighboring pixels in a CCD during readout. This process is performed prior to digitization in the CCD chip using dedicated serial and parallel register control. The two main benefits of binning are improved signal-to-noise ratio (SNR) and the ability to increase frame rates, albeit at the cost of reduced spatial resolution.

      Bioconservatism (a portmanteau of biology and conservatism) is a stance of hesitancy and skepticism regarding radical technological advances, especially those that seek to modify or enhance the human condition. Bioconservatism is characterized by a belief that technological trends in today’s society risk compromising human dignity, and by opposition to movements and technologies including transhumanism, human genetic modification, «strong» artificial intelligence, and the technological singularity. Many bioconservatives also oppose the use of technologies such as life extension and preimplantation genetic screening168,169.

      Biometrics is a people recognition system, one or more physical or behavioral traits170,171.

      Black box is a description of some deep learning system. They take an input and provide an output, but the calculations that occur in between are not easy for humans to interpret172,173.

      Blackboard system is an artificial intelligence approach based on the blackboard architectural model, where a common knowledge base, the «blackboard», is iteratively updated by a diverse group of specialist knowledge sources, starting with a problem specification and ending with a solution. Each knowledge source updates the blackboard with a partial solution when its internal constraints match the blackboard state. In this way, the specialists work together to solve the problem174.

      BLEU (Bilingual Evaluation Understudy) is a text quality evaluation algorithm between 0.0 and 1.0, inclusive, indicating the quality of a translation between two human languages (for example, between English and Russian). A BLEU score of 1.0 indicates a perfect translation; a BLEU score of 0.0 indicates a terrible translation175.

      Blockchain is algorithms and protocols for decentralized storage and processing of transactions structured as a sequence of linked blocks without the possibility of their subsequent change176.

      Boltzmann machine (also stochastic Hopfield network with hidden units) is a type of stochastic recurrent neural network and Markov random field. Boltzmann machines can be seen as the stochastic, generative counterpart of Hopfield networks177.

      Boolean neural network is an artificial neural network approach which only consists of Boolean neurons (and, or, not). Such an approach reduces the use of memory space and computation time. It can be implemented to the programmable circuits such as FPGA (Field-Programmable Gate Array or Integrated circuit).

      Boolean satisfiability problem (also propositional satisfiability problem; abbreviated SATISFIABILITY or SAT) is the problem of determining if there exists an interpretation that satisfies a given Boolean formula. In other words, it asks whether the variables of a given Boolean formula can be consistently replaced by the values TRUE or FALSE in such a way that the formula evaluates to TRUE. If this is the case, the formula is called satisfiable. On the other hand, if no such assignment exists, the function expressed by the formula is FALSE for all possible variable assignments and the formula is unsatisfiable178.

      Boosting is a Machine Learning ensemble meta-algorithm for primarily reducing bias and variance in supervised learning, and a family of Machine Learning algorithms that convert weak learners to strong ones179.

      Bounding Box commonly used in image or video tagging; this is an imaginary box drawn on visual information. The contents of the box are labeled to help a model recognize it as a distinct type of object.

      Brain technology (also self-learning know-how system) is a technology that employs the latest findings in neuroscience. The term was first introduced by the Artificial Intelligence Laboratory in Zurich, Switzerland, in the context of the ROBOY project. Brain Technology can be employed in robots, know-how management systems and any other application with self-learning capabilities.

Скачать книгу


<p>162</p>

Bigram [Электронный ресурс] https://en.wikipedia.org URL: https://en.wikipedia.org/wiki/Bigram (дата обращения: 28.03.2023)

<p>163</p>

Регрессионная модель бинарного выбора [Электронный ресурс] https://www.statmethods.ru URL: https://www.statmethods.ru/statistics-metody/logisticheskaya-regressiya/ (дата обращения: 29.06.2023)

<p>164</p>

Binary classification [Электронный ресурс] https://www.primeclasses.in URL: https://www.primeclasses.in/glossary/data-science-course/machine-learning/binary-classification (дата обращения: 09.04.2023)

<p>165</p>

Binary format [Электронный ресурс] www.umich.edu URL: https://www.icpsr.umich.edu/web/ICPSR/cms/2042#B (дата обращения: 07.07.2022)

<p>166</p>

Binary number [Электронный ресурс] www.umich.edu URL: https://www.icpsr.umich.edu/web/ICPSR/cms/2042#B (дата обращения: 07.07.2022)

<p>167</p>

Binary tree [Электронный ресурс] https://habr.com URL: https://habr.com/ru/post/267855/ (дата обращения: 31.01.2022)

<p>168</p>

Bioconservatism [Электронный ресурс] https://en.wikipedia.org URL: https://en.wikipedia.org/wiki/Bioconservatism (дата обращения: 07.07.2022)

<p>169</p>

.Bioconservatism [Электронный ресурс] www.wise-geek.com URL: https://www.wise-geek.com/what-is-bioconservatism.htm (дата обращения: 07.07.2022)

<p>170</p>

Biometrics [Электронный ресурс] https://recfaces.com URL: https://recfaces.com/articles/types-of-biometrics (дата обращения: 28.03.2023)

<p>171</p>

Биометрия [Электронный ресурс] https://ru.wikipedia.org URL: https://ru.wikipedia.org/wiki/Биометрия (дата обращения: 28.03.2023)

<p>172</p>

Black Box [Электронный ресурс] https://www.linkedin.com URL: https://www.linkedin.com/pulse/25-artificial-intelligence-terms-you-need-know-today-jair-ribeiro (дата обращения: 10.07.2023)

<p>173</p>

Чёрный ящик [Электронный ресурс] https://ru.wikipedia.org URL: https://ru.wikipedia.org/wiki/Чёрный_ящик (дата обращения: 10.07.2023)

<p>174</p>

Blackboard system [Электронный ресурс] https://studme.org URL: https://studme.org/282503/informatika/sistemy_ispolzuyuschie_printsip_klassnoy_doski (дата обращения: 19.02.2022)

<p>175</p>

BLEU (Bilingual Evaluation Understudy) [Электронный ресурс] https://developers.google.com URL: https://developers.google.com/machine-learning/glossary/language (дата обращения: 27.03.2023)

<p>176</p>

Blockchain [Электронный ресурс] https://books.google.ru URL: https://books.google.ru/books?id= -w2DEAAAQBAJ&pg=PT301&lpg= PT301&dq=algorithms+and +protocols+for+decentralized+storage+and+processing+of +transactions +structured+as+a+sequence+of+linked+blocks+without+the+ possibility +of+their+subsequent+change&source =bl&ots=ruX5Ow4F-g&sig= ACfU3U0fmUw6tcXOQoRbPdWNAfwf5AFYWQ&hl =ru&sa=X&ved=2ahUKEwjmpsrjx_ 79AhWDmIsKHQqDBvMQ6AF6BAgrEAM#v= onepage&q=algorithms%20and %20protocols%20for %20decentralized%20storage%20and%20processing %20of%20transactions %20structured%20as%20a%20sequence %20of%20linked%20blocks %20without%20the %20possibility%20of%20their% 20subsequent %20change&f=false (дата обращения: 28.03.2023)

<p>177</p>

Boltzmann machine [Электронный ресурс] https://dic.academic.ru URL: https://dic.academic.ru/dic.nsf/ruwiki/1828062 (дата обращения: 04.02.2022)

<p>178</p>

Boolean satisfiability problem A. de Carvalho M.C. Fairhurst D.L. Bisset, An integrated Boolean neural network for pattern classification. Pattern Recognition Letters Volume 15, Issue 8, August 1994, Pages 807—813 (дата обращения: 10.02.2022)

<p>179</p>

Boosting [Электронный ресурс] https://en.wikipedia.org URL: https://en.wikipedia.org/wiki/Boosting_(machine_learning) (дата обращения: 28.03.2023)