Пређи на садржај

Машинско учење — разлика између измена

С Википедије, слободне енциклопедије
Садржај обрисан Садржај додат
м нормативна контрола
.
Ред 1: Ред 1:
{{Short description|Проучавање алгоритама који се аутоматски побољшавају кроз искуство}}
'''Машинско учење''' је подобласт [[вјештачка интелигенција|вјештачке интелигенције]] чији је циљ конструисање [[алгоритам]]а и рачунарских система који су способни да се адаптирају на аналогне нове ситуације и уче на бази искуства. Развијене су различите технике учења за извршавање различитих задатака. Прве које су биле предмет истраживања, тичу се надгледаног учења за дискреционо доношење одлука, надгледаног учења за континуирано предвиђање и појачано учење за секвенционално доношење одлука, као и ненадгледано учење.


До сада најбоље схваћен од свих наведених задатака је одлучивање преко једног покушаја ({{јез-енгл|one-shot learning}}). [[Рачунар]]у је дат опис једног објекта (догађаја или ситуације) и од њега се очекује да као резултат избаци класификацију тог објекта. На примјер, програм за [[препознавање алфанумеричких знакова]] као улазну вриједност има дигитализовану слику неког алфанумеричког знака и као резултат треба да избаци његово име.
'''Машинско учење''' ({{јез-енг|Machine learning, ML}}) је подобласт [[вјештачка интелигенција|вештачке интелигенције]] чији је циљ конструисање [[Computational statistics|статистичких алгоритама]] и рачунарских система који су способни да се адаптирају на аналогне нове ситуације и уче на бази искуства. Развијене су различите технике учења за извршавање различитих задатака. Прве које су биле предмет истраживања, тичу се надгледаног учења за дискреционо доношење одлука, надгледаног учења за континуирано предвиђање и појачано учење за секвенционално доношење одлука, као и ненадгледано учење. До сада најбоље схваћен од свих наведених задатака је одлучивање преко једног покушаја ({{јез-енгл|one-shot learning}}). [[Рачунар]]у је дат опис једног објекта (догађаја или ситуације) и од њега се очекује да као резултат избаци класификацију тог објекта. На примјер, програм за [[препознавање алфанумеричких знакова]] као улазну вриједност има дигитализовану слику неког алфанумеричког знака и као резултат треба да избаци његово име.

Алгоритми машинског учења могу да уче из [[data|података]] и [[generalize|генерализују]] на невидљиве податке, и на тај начин обављају [[Task (computing)|задатке]] без експлицитних [[Machine code|упутстава]].{{refn|The definition "without being explicitly programmed" is often attributed to [[Arthur Samuel (computer scientist)|Arthur Samuel]], who coined the term "machine learning" in 1959, but the phrase is not found verbatim in this publication, and may be a [[paraphrase]] that appeared later. Confer "Paraphrasing Arthur Samuel (1959), the question is: How can computers learn to solve problems without being explicitly programmed?" in {{Cite conference |chapter=Automated Design of Both the Topology and Sizing of Analog Electrical Circuits Using Genetic Programming |conference=Artificial Intelligence in Design '96 |last1=Koza |first1=John R. |last2=Bennett |first2=Forrest H. |last3=Andre |first3=David |last4=Keane |first4=Martin A. |title=Artificial Intelligence in Design '96 |date=1996 |publisher=Springer, Dordrecht |pages=151–170 |language=en |doi=10.1007/978-94-009-0279-4_9 |isbn=978-94-010-6610-5 }}}} Недавно су [[artificial neural network|вештачке неуронске мреже]] успеле да надмаше многе претходне приступе у погледу перформанси.<ref name="ibm">{{Cite web |title=What is Machine Learning? |url=https://www.ibm.com/topics/machine-learning |access-date=2023-06-27 |website=IBM |language=en-us}}</ref><ref name=":6">{{Cite web |last=Zhou |first=Victor |date=2019-12-20 |title=Machine Learning for Beginners: An Introduction to Neural Networks |url=https://towardsdatascience.com/machine-learning-for-beginners-an-introduction-to-neural-networks-d49f22d238f9 |url-status=live |access-date=2021-08-15 |website=Medium |language=en |archive-date=2022-03-09 |archive-url=https://web.archive.org/web/20220309053518/https://towardsdatascience.com/machine-learning-for-beginners-an-introduction-to-neural-networks-d49f22d238f9 }}</ref> Приступи машинском учењу су примењени у многим областима укључујући [[natural language processing|обраду природног језика]], [[computer vision|компјутерски вид]], [[speech recognition|препознавање говора]], [[email filtering|филтрирање електронске поште]], [[agriculture|пољопривреду]] и медицину.<ref name="tvt">{{Cite journal |last1=Hu |first1=Junyan |last2=Niu |first2=Hanlin |last3=Carrasco |first3=Joaquin |last4=Lennox |first4=Barry |last5=Arvin |first5=Farshad |date=2020 |title=Voronoi-Based Multi-Robot Autonomous Exploration in Unknown Environments via Deep Reinforcement Learning |journal=IEEE Transactions on Vehicular Technology |volume=69 |issue=12 |pages=14413–14423 |doi=10.1109/tvt.2020.3034800 |s2cid=228989788 |issn=0018-9545 |doi-access=free }}</ref><ref name=":7">{{cite journal |last1=Yoosefzadeh-Najafabadi|first1=Mohsen |last2=Hugh |first2=Earl |last3=Tulpan |first3=Dan |last4=Sulik |first4=John |last5=Eskandari |first5=Milad |title=Application of Machine Learning Algorithms in Plant Breeding: Predicting Yield From Hyperspectral Reflectance in Soybean? |journal=Front. Plant Sci. |volume=11 |year=2021 |pages=624273|doi=10.3389/fpls.2020.624273 |pmid=33510761 |pmc=7835636 |doi-access=free }}</ref> ML је познато по својој примени на пословне проблеме под називом [[predictive analytics|предиктивна аналитика]]. Иако није свако машинско учење [[Statistics|статистички]] засновано, [[computational statistics|рачунарска статистика]] је важан извор метода у овој области.

Математичке основе машинског учења су обезбеђене методама [[mathematical optimization|математичке оптимизације]] (математичко програмирање). [[Истраживање података]] је сродна (паралелна) област проучавања, која се фокусира на [[exploratory data analysis|истраживачку анализу података]] кроз [[unsupervised learning|учење без надзора]].{{refn|Machine learning and pattern recognition "can be viewed as two facets of the same field".<ref name="bishop2006" />{{rp|vii}}}}<ref name=":9">{{cite journal |last=Friedman |first=Jerome H. |author-link = Jerome H. Friedman|title=Data Mining and Statistics: What's the connection? |journal=Computing Science and Statistics |volume=29 |issue=1 |year=1998 |pages=3–9}}</ref> Са теоријске тачке гледишта, [[Probably approximately correct learning|вероватно приближно тачно учење]] пружа оквир за описивање машинског учења.

== Историја ==
{{rut}}
The term ''machine learning'' was coined in 1959 by [[Arthur Samuel (computer scientist)|Arthur Samuel]], an [[IBM]] employee and pioneer in the field of [[computer gaming]] and [[artificial intelligence]].<ref name="Samuel">{{Cite journal|last=Samuel|first=Arthur|date=1959|title=Some Studies in Machine Learning Using the Game of Checkers|journal=IBM Journal of Research and Development|volume=3|issue=3|pages=210–229|doi=10.1147/rd.33.0210|citeseerx=10.1.1.368.2254|s2cid=2126705 }}</ref><ref name=":8">R. Kohavi and F. Provost, "Glossary of terms", Machine Learning, vol. 30, no. 2–3, pp. 271–274, 1998.</ref> The synonym ''self-teaching computers'' was also used in this time period.<ref name=cyberthreat>{{cite news |last1=Gerovitch |first1=Slava |title=How the Computer Got Its Revenge on the Soviet Union |url=https://nautil.us/issue/23/dominoes/how-the-computer-got-its-revenge-on-the-soviet-union |access-date=19 September 2021 |work=Nautilus |date=9 April 2015 |archive-date=22 September 2021 |archive-url=https://web.archive.org/web/20210922175839/https://nautil.us/issue/23/Dominoes/how-the-computer-got-its-revenge-on-the-soviet-union |url-status=dead }}</ref><ref>{{cite journal |last1=Lindsay |first1=Richard P. |title=The Impact of Automation On Public Administration |journal=Western Political Quarterly |date=1 September 1964 |volume=17 |issue=3 |pages=78–81 |doi=10.1177/106591296401700364 |s2cid=154021253 |url=https://journals.sagepub.com/doi/10.1177/106591296401700364 |access-date=6 October 2021 |language=en |issn=0043-4078 |archive-date=6 October 2021 |archive-url=https://web.archive.org/web/20211006190841/https://journals.sagepub.com/doi/10.1177/106591296401700364 |url-status=live }}</ref>

Although the earliest machine learning model was introduced in the 1950s when [[Arthur Samuel (computer scientist)|Arthur Samuel]] invented a [[Computer program|program]] that calculated the winning chance in checkers for each side, the history of machine learning roots back to decades of human desire and effort to study human cognitive processes.<ref name=":02">{{Cite web |title=History and Evolution of Machine Learning: A Timeline |url=https://www.techtarget.com/whatis/A-Timeline-of-Machine-Learning-History |access-date=2023-12-08 |website=WhatIs |language=en}}</ref> In 1949, Canadian psychologist [[Donald O. Hebb|Donald Hebb]] published the book ''[[Organization of Behavior|The Organization of Behavior]]'', in which he introduced a [[Hebbian theory|theoretical neural structure]] formed by certain interactions among [[nerve cells]].<ref>{{Cite journal |last=Milner |first=Peter M. |date=1993 |title=The Mind and Donald O. Hebb |url=https://www.jstor.org/stable/24941344 |journal=Scientific American |volume=268 |issue=1 |pages=124–129 |doi=10.1038/scientificamerican0193-124 |jstor=24941344 |pmid=8418480 |bibcode=1993SciAm.268a.124M |issn=0036-8733}}</ref> Hebb's model of [[neuron]]s interacting with one another set a groundwork for how AIs and machine learning algorithms work under nodes, or [[artificial neuron]]s used by computers to communicate data.<ref name=":02" /> Other researchers who have studied human cognitive systems contributed to the modern machine learning technologies as well, including logician [[Walter Pitts]] and [[Warren Sturgis McCulloch|Warren McCulloch]], who proposed the early mathematical models of neural networks to come up with algorithms that mirror human thought processes.<ref name=":02" />

By the early 1960s an experimental "learning machine" with [[punched tape]] memory, called Cybertron, had been developed by [[Raytheon Company]] to analyze [[sonar]] signals, [[Electrocardiography|electrocardiograms]], and speech patterns using rudimentary [[reinforcement learning]]. It was repetitively "trained" by a human operator/teacher to recognize patterns and equipped with a "[[goof]]" button to cause it to re-evaluate incorrect decisions.<ref>"Science: The Goof Button", [[Time (magazine)]], 18 August 1961.
</ref> A representative book on research into machine learning during the 1960s was Nilsson's book on Learning Machines, dealing mostly with machine learning for pattern classification.<ref>Nilsson N. Learning Machines, McGraw Hill, 1965.</ref> Interest related to pattern recognition continued into the 1970s, as described by Duda and Hart in 1973.<ref>Duda, R., Hart P. Pattern Recognition and Scene Analysis, Wiley Interscience, 1973</ref> In 1981 a report was given on using teaching strategies so that an [[artificial neural network]] learns to recognize 40 characters (26 letters, 10 digits, and 4 special symbols) from a computer terminal.<ref>S. Bozinovski "Teaching space: A representation concept for adaptive pattern classification" COINS Technical Report No. 81-28, Computer and Information Science Department, University of Massachusetts at Amherst, MA, 1981. https://web.cs.umass.edu/publication/docs/1981/UM-CS-1981-028.pdf {{Webarchive|url=https://web.archive.org/web/20210225070218/https://web.cs.umass.edu/publication/docs/1981/UM-CS-1981-028.pdf |date=2021-02-25 }}</ref>

[[Tom M. Mitchell]] provided a widely quoted, more formal definition of the algorithms studied in the machine learning field: "A computer program is said to learn from experience ''E'' with respect to some class of tasks ''T'' and performance measure ''P'' if its performance at tasks in ''T'', as measured by ''P'', improves with experience ''E''."<ref name="Mitchell-1997">{{cite book |author=Mitchell, T. |title=Machine Learning |publisher=McGraw Hill |isbn= 978-0-07-042807-2 |pages=2 |year=1997}}</ref> This definition of the tasks in which machine learning is concerned offers a fundamentally [[operational definition]] rather than defining the field in cognitive terms. This follows [[Alan Turing]]'s proposal in his paper "[[Computing Machinery and Intelligence]]", in which the question "Can machines think?" is replaced with the question "Can machines do what we (as thinking entities) can do?".<ref>{{Citation |chapter-url=http://eprints.ecs.soton.ac.uk/12954/ |first=Stevan |last=Harnad |author-link=Stevan Harnad |year=2008 |chapter=The Annotation Game: On Turing (1950) on Computing, Machinery, and Intelligence |editor1-last=Epstein |editor1-first=Robert |editor2-last=Peters |editor2-first=Grace |title=The Turing Test Sourcebook: Philosophical and Methodological Issues in the Quest for the Thinking Computer |pages=23–66 |publisher=Kluwer |isbn=9781402067082 |access-date=2012-12-11 |archive-date=2012-03-09 |archive-url=https://web.archive.org/web/20120309113922/http://eprints.ecs.soton.ac.uk/12954/ }}</ref>

Modern-day machine learning has two objectives. One is to classify data based on models which have been developed; the other purpose is to make predictions for future outcomes based on these models. A hypothetical algorithm specific to classifying data may use computer vision of moles coupled with supervised learning in order to train it to classify the cancerous moles. A machine learning algorithm for stock trading may inform the trader of future potential predictions.<ref>{{Cite web|date=2020-12-08|title=Introduction to AI Part 1|url=https://edzion.com/2020/12/09/introduction-to-ai-part-1/|access-date=2020-12-09|website=Edzion|language=en|archive-date=2021-02-18|archive-url=https://web.archive.org/web/20210218005157/https://edzion.com/2020/12/09/introduction-to-ai-part-1/|url-status=live}}</ref>


== Види још ==
== Види још ==
Ред 8: Ред 25:
* [[Формалистичка теорија процеса учења природних језика]]
* [[Формалистичка теорија процеса учења природних језика]]
* [[Спознаја|Когниција]]
* [[Спознаја|Когниција]]

== Референце ==
{{Reflist|refs=
<ref name="bishop2006">{{citation|first= C. M. |last= Bishop |author-link=Christopher M. Bishop |year=2006 |title=Pattern Recognition and Machine Learning |publisher=Springer |isbn=978-0-387-31073-2}}</ref>

}}


== Литература ==
== Литература ==
{{Refbegin|2}}-{
{{Refbegin|30em}}
* {{cite book | last = Domingos | first = Pedro | author-link = Pedro Domingos | title = The Master Algorithm: How the Quest for the Ultimate Learning Machine Will Remake Our World | date = September 22, 2015 | publisher = [[Basic Books]] | isbn = 978-0465065707 }}
* {{cite book|last=Nilsson|first=Nils|author-link=Nils Nilsson (researcher)|year=1998|title=Artificial Intelligence: A New Synthesis|url=https://archive.org/details/artificialintell0000nils|url-access=registration|publisher=Morgan Kaufmann|isbn=978-1-55860-467-4|access-date=18 November 2019|archive-date=26 July 2020|archive-url=https://web.archive.org/web/20200726131654/https://archive.org/details/artificialintell0000nils|url-status=live}}
* {{cite book|first1=David|last1=Poole|first2=Alan|last2=Mackworth|author2-link=Alan Mackworth|first3=Randy|last3=Goebel|year=1998|title=Computational Intelligence: A Logical Approach|publisher=Oxford University Press|location=New York|isbn=978-0-19-510270-3|url=https://archive.org/details/computationalint00pool|access-date=22 August 2020|archive-date=26 July 2020|archive-url=https://web.archive.org/web/20200726131436/https://archive.org/details/computationalint00pool|url-status=live}}
* Nils J. Nilsson, ''[https://ai.stanford.edu/people/nilsson/mlbook.html Introduction to Machine Learning] {{Webarchive|url=https://web.archive.org/web/20190816182600/http://ai.stanford.edu/people/nilsson/mlbook.html |date=2019-08-16 }}''.
* [[David J. C. MacKay]]. ''[http://www.inference.phy.cam.ac.uk/mackay/itila/book.html Information Theory, Inference, and Learning Algorithms] {{Webarchive|url=https://web.archive.org/web/20160217105359/http://www.inference.phy.cam.ac.uk/mackay/itila/book.html |date=2016-02-17 }}'' Cambridge: Cambridge University Press, 2003. {{ISBN|0-521-64298-1}}
* Stuart Russell & Peter Norvig, (2009). ''[http://aima.cs.berkeley.edu/ Artificial Intelligence – A Modern Approach] {{Webarchive|url=https://web.archive.org/web/20110228023805/http://aima.cs.berkeley.edu/ |date=2011-02-28 }}''. Pearson, {{ISBN|9789332543515}}.
* Kevin P. Murphy (2021). ''[https://probml.github.io/pml-book/book1.html Probabilistic Machine Learning: An Introduction] {{Webarchive|url=https://web.archive.org/web/20210411153246/https://probml.github.io/pml-book/book1.html |date=2021-04-11 }}'', MIT Press.
* Sergios Theodoridis, Konstantinos Koutroumbas (2009) "Pattern Recognition", 4th Edition, Academic Press. {{ISBN|978-1-59749-272-0}}.
* Sergios Theodoridis, Konstantinos Koutroumbas (2009) "Pattern Recognition", 4th Edition, Academic Press. {{ISBN|978-1-59749-272-0}}.
* Ethem Alpaydın (2004) ''Introduction to Machine Learning (Adaptive Computation and Machine Learning)'', MIT Press, {{ISBN|978-0-262-01211-9}}
* Ethem Alpaydın (2004) ''Introduction to Machine Learning (Adaptive Computation and Machine Learning)'', MIT Press, {{ISBN|978-0-262-01211-9}}
Ред 25: Ред 55:
* Huang T.-M., Kecman V., Kopriva I. (2006), [http://learning-from-data.com Kernel Based Algorithms for Mining Huge Data Sets, Supervised, Semi-supervised, and Unsupervised Learning], Springer-Verlag, Berlin, Heidelberg, 260 pp. 96 illus., Hardcover, {{ISBN|978-3-540-31681-7}}.
* Huang T.-M., Kecman V., Kopriva I. (2006), [http://learning-from-data.com Kernel Based Algorithms for Mining Huge Data Sets, Supervised, Semi-supervised, and Unsupervised Learning], Springer-Verlag, Berlin, Heidelberg, 260 pp. 96 illus., Hardcover, {{ISBN|978-3-540-31681-7}}.
* KECMAN Vojislav (2001), [http://support-vector.ws Learning and Soft Computing, Support Vector Machines, Neural Networks and Fuzzy Logic Models], The MIT Press, Cambridge, MA, 608 pp., 268 illus., {{ISBN|978-0-262-11255-0}}.
* KECMAN Vojislav (2001), [http://support-vector.ws Learning and Soft Computing, Support Vector Machines, Neural Networks and Fuzzy Logic Models], The MIT Press, Cambridge, MA, 608 pp., 268 illus., {{ISBN|978-0-262-11255-0}}.
* MacKay, D.J.C. (2003). ''[http://www.inference.phy.cam.ac.uk/mackay/itila/ Information Theory, Inference, and Learning Algorithms]'', Cambridge University Press. {{ISBN|978-0-521-64298-9}}.
* Ian H. Witten and Eibe Frank ''Data Mining: Practical machine learning tools and techniques'' Morgan Kaufmann {{ISBN|978-0-12-088407-0}}.
* Ian H. Witten and Eibe Frank ''Data Mining: Practical machine learning tools and techniques'' Morgan Kaufmann {{ISBN|978-0-12-088407-0}}.
* Sholom Weiss and Casimir Kulikowski (1991). ''Computer Systems That Learn'', Morgan Kaufmann. {{ISBN|978-1-55860-065-2}}.
* Sholom Weiss and Casimir Kulikowski (1991). ''Computer Systems That Learn'', Morgan Kaufmann. {{ISBN|978-1-55860-065-2}}.
Ред 31: Ред 60:
* Trevor Hastie, Robert Tibshirani and Jerome Friedman (2001). ''[https://web.archive.org/web/20091110212529/http://www-stat.stanford.edu/~tibs/ElemStatLearn/ The Elements of Statistical Learning]'', Springer. {{ISBN|978-0-387-95284-0}}.
* Trevor Hastie, Robert Tibshirani and Jerome Friedman (2001). ''[https://web.archive.org/web/20091110212529/http://www-stat.stanford.edu/~tibs/ElemStatLearn/ The Elements of Statistical Learning]'', Springer. {{ISBN|978-0-387-95284-0}}.
* Vladimir Vapnik (1998). ''Statistical Learning Theory''. Wiley-Interscience, {{ISBN|978-0-471-03003-4}}.
* Vladimir Vapnik (1998). ''Statistical Learning Theory''. Wiley-Interscience, {{ISBN|978-0-471-03003-4}}.
}-{{Refend}}
{{Refend}}

== Спољашње везе ==
{{Commons category|Machine learning}}
*[https://web.archive.org/web/20171230081341/http://machinelearning.org/ International Machine Learning Society]
*[https://mloss.org/ mloss] is an academic database of open-source machine learning software.


{{нормативна контрола}}
{{нормативна контрола}}

Верзија на датум 9. март 2024. у 19:44

Машинско учење (енгл. Machine learning, ML) је подобласт вештачке интелигенције чији је циљ конструисање статистичких алгоритама и рачунарских система који су способни да се адаптирају на аналогне нове ситуације и уче на бази искуства. Развијене су различите технике учења за извршавање различитих задатака. Прве које су биле предмет истраживања, тичу се надгледаног учења за дискреционо доношење одлука, надгледаног учења за континуирано предвиђање и појачано учење за секвенционално доношење одлука, као и ненадгледано учење. До сада најбоље схваћен од свих наведених задатака је одлучивање преко једног покушаја (енгл. one-shot learning). Рачунару је дат опис једног објекта (догађаја или ситуације) и од њега се очекује да као резултат избаци класификацију тог објекта. На примјер, програм за препознавање алфанумеричких знакова као улазну вриједност има дигитализовану слику неког алфанумеричког знака и као резултат треба да избаци његово име.

Алгоритми машинског учења могу да уче из података и генерализују на невидљиве податке, и на тај начин обављају задатке без експлицитних упутстава.[1] Недавно су вештачке неуронске мреже успеле да надмаше многе претходне приступе у погледу перформанси.[2][3] Приступи машинском учењу су примењени у многим областима укључујући обраду природног језика, компјутерски вид, препознавање говора, филтрирање електронске поште, пољопривреду и медицину.[4][5] ML је познато по својој примени на пословне проблеме под називом предиктивна аналитика. Иако није свако машинско учење статистички засновано, рачунарска статистика је важан извор метода у овој области.

Математичке основе машинског учења су обезбеђене методама математичке оптимизације (математичко програмирање). Истраживање података је сродна (паралелна) област проучавања, која се фокусира на истраживачку анализу података кроз учење без надзора.[7][8] Са теоријске тачке гледишта, вероватно приближно тачно учење пружа оквир за описивање машинског учења.

Историја

The term machine learning was coined in 1959 by Arthur Samuel, an IBM employee and pioneer in the field of computer gaming and artificial intelligence.[9][10] The synonym self-teaching computers was also used in this time period.[11][12]

Although the earliest machine learning model was introduced in the 1950s when Arthur Samuel invented a program that calculated the winning chance in checkers for each side, the history of machine learning roots back to decades of human desire and effort to study human cognitive processes.[13] In 1949, Canadian psychologist Donald Hebb published the book The Organization of Behavior, in which he introduced a theoretical neural structure formed by certain interactions among nerve cells.[14] Hebb's model of neurons interacting with one another set a groundwork for how AIs and machine learning algorithms work under nodes, or artificial neurons used by computers to communicate data.[13] Other researchers who have studied human cognitive systems contributed to the modern machine learning technologies as well, including logician Walter Pitts and Warren McCulloch, who proposed the early mathematical models of neural networks to come up with algorithms that mirror human thought processes.[13]

By the early 1960s an experimental "learning machine" with punched tape memory, called Cybertron, had been developed by Raytheon Company to analyze sonar signals, electrocardiograms, and speech patterns using rudimentary reinforcement learning. It was repetitively "trained" by a human operator/teacher to recognize patterns and equipped with a "goof" button to cause it to re-evaluate incorrect decisions.[15] A representative book on research into machine learning during the 1960s was Nilsson's book on Learning Machines, dealing mostly with machine learning for pattern classification.[16] Interest related to pattern recognition continued into the 1970s, as described by Duda and Hart in 1973.[17] In 1981 a report was given on using teaching strategies so that an artificial neural network learns to recognize 40 characters (26 letters, 10 digits, and 4 special symbols) from a computer terminal.[18]

Tom M. Mitchell provided a widely quoted, more formal definition of the algorithms studied in the machine learning field: "A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P if its performance at tasks in T, as measured by P, improves with experience E."[19] This definition of the tasks in which machine learning is concerned offers a fundamentally operational definition rather than defining the field in cognitive terms. This follows Alan Turing's proposal in his paper "Computing Machinery and Intelligence", in which the question "Can machines think?" is replaced with the question "Can machines do what we (as thinking entities) can do?".[20]

Modern-day machine learning has two objectives. One is to classify data based on models which have been developed; the other purpose is to make predictions for future outcomes based on these models. A hypothetical algorithm specific to classifying data may use computer vision of moles coupled with supervised learning in order to train it to classify the cancerous moles. A machine learning algorithm for stock trading may inform the trader of future potential predictions.[21]

Види још

Референце

  1. ^ The definition "without being explicitly programmed" is often attributed to Arthur Samuel, who coined the term "machine learning" in 1959, but the phrase is not found verbatim in this publication, and may be a paraphrase that appeared later. Confer "Paraphrasing Arthur Samuel (1959), the question is: How can computers learn to solve problems without being explicitly programmed?" in Koza, John R.; Bennett, Forrest H.; Andre, David; Keane, Martin A. (1996). „Automated Design of Both the Topology and Sizing of Analog Electrical Circuits Using Genetic Programming”. Artificial Intelligence in Design '96. Artificial Intelligence in Design '96 (на језику: енглески). Springer, Dordrecht. стр. 151—170. ISBN 978-94-010-6610-5. doi:10.1007/978-94-009-0279-4_9. 
  2. ^ „What is Machine Learning?”. IBM (на језику: енглески). Приступљено 2023-06-27. 
  3. ^ Zhou, Victor (2019-12-20). „Machine Learning for Beginners: An Introduction to Neural Networks”. Medium (на језику: енглески). Архивирано из оригинала 2022-03-09. г. Приступљено 2021-08-15. 
  4. ^ Hu, Junyan; Niu, Hanlin; Carrasco, Joaquin; Lennox, Barry; Arvin, Farshad (2020). „Voronoi-Based Multi-Robot Autonomous Exploration in Unknown Environments via Deep Reinforcement Learning”. IEEE Transactions on Vehicular Technology. 69 (12): 14413—14423. ISSN 0018-9545. S2CID 228989788. doi:10.1109/tvt.2020.3034800Слободан приступ. 
  5. ^ Yoosefzadeh-Najafabadi, Mohsen; Hugh, Earl; Tulpan, Dan; Sulik, John; Eskandari, Milad (2021). „Application of Machine Learning Algorithms in Plant Breeding: Predicting Yield From Hyperspectral Reflectance in Soybean?”. Front. Plant Sci. 11: 624273. PMC 7835636Слободан приступ. PMID 33510761. doi:10.3389/fpls.2020.624273Слободан приступ. 
  6. ^ Bishop, C. M. (2006), Pattern Recognition and Machine Learning, Springer, ISBN 978-0-387-31073-2 
  7. ^ Machine learning and pattern recognition "can be viewed as two facets of the same field".[6]:vii
  8. ^ Friedman, Jerome H. (1998). „Data Mining and Statistics: What's the connection?”. Computing Science and Statistics. 29 (1): 3—9. 
  9. ^ Samuel, Arthur (1959). „Some Studies in Machine Learning Using the Game of Checkers”. IBM Journal of Research and Development. 3 (3): 210—229. CiteSeerX 10.1.1.368.2254Слободан приступ. S2CID 2126705. doi:10.1147/rd.33.0210. 
  10. ^ R. Kohavi and F. Provost, "Glossary of terms", Machine Learning, vol. 30, no. 2–3, pp. 271–274, 1998.
  11. ^ Gerovitch, Slava (9. 4. 2015). „How the Computer Got Its Revenge on the Soviet Union”. Nautilus. Архивирано из оригинала 22. 9. 2021. г. Приступљено 19. 9. 2021. 
  12. ^ Lindsay, Richard P. (1. 9. 1964). „The Impact of Automation On Public Administration”. Western Political Quarterly (на језику: енглески). 17 (3): 78—81. ISSN 0043-4078. S2CID 154021253. doi:10.1177/106591296401700364. Архивирано из оригинала 6. 10. 2021. г. Приступљено 6. 10. 2021. 
  13. ^ а б в „History and Evolution of Machine Learning: A Timeline”. WhatIs (на језику: енглески). Приступљено 2023-12-08. 
  14. ^ Milner, Peter M. (1993). „The Mind and Donald O. Hebb”. Scientific American. 268 (1): 124—129. Bibcode:1993SciAm.268a.124M. ISSN 0036-8733. JSTOR 24941344. PMID 8418480. doi:10.1038/scientificamerican0193-124. 
  15. ^ "Science: The Goof Button", Time (magazine), 18 August 1961.
  16. ^ Nilsson N. Learning Machines, McGraw Hill, 1965.
  17. ^ Duda, R., Hart P. Pattern Recognition and Scene Analysis, Wiley Interscience, 1973
  18. ^ S. Bozinovski "Teaching space: A representation concept for adaptive pattern classification" COINS Technical Report No. 81-28, Computer and Information Science Department, University of Massachusetts at Amherst, MA, 1981. https://web.cs.umass.edu/publication/docs/1981/UM-CS-1981-028.pdf Архивирано 2021-02-25 на сајту Wayback Machine
  19. ^ Mitchell, T. (1997). Machine Learning. McGraw Hill. стр. 2. ISBN 978-0-07-042807-2. 
  20. ^ Harnad, Stevan (2008), „The Annotation Game: On Turing (1950) on Computing, Machinery, and Intelligence”, Ур.: Epstein, Robert; Peters, Grace, The Turing Test Sourcebook: Philosophical and Methodological Issues in the Quest for the Thinking Computer, Kluwer, стр. 23—66, ISBN 9781402067082, Архивирано из оригинала 2012-03-09. г., Приступљено 2012-12-11 
  21. ^ „Introduction to AI Part 1”. Edzion (на језику: енглески). 2020-12-08. Архивирано из оригинала 2021-02-18. г. Приступљено 2020-12-09. 

Литература

Спољашње везе