User:Spanini/sandbox: Difference between revisions
Conference edits |
|||
| Line 5: | Line 5: | ||
=== '''L* Algorithm''' === |
=== '''L* Algorithm''' === |
||
Angluin has written highly cited papers on [[computational learning theory]], particularly in the context of learning [[regular language]] sets from membership and equivalence queries using the L* algorithm<ref>{{Cite journal |last=Grinchtein |first=Olga |last2=Jonsson |first2=Bengt |last3=Leucker |first3=Martin |date=2010-10 |title=Learning of event-recording automata |url=https://linkinghub.elsevier.com/retrieve/pii/S0304397510003944 |journal=Theoretical Computer Science |language=en |volume=411 |issue=47 |pages=4029–4054 |doi=10.1016/j.tcs.2010.07.008}}</ref>. |
Angluin has written highly cited papers on [[computational learning theory]], particularly in the context of learning [[regular language]] sets from membership and equivalence queries using the L* algorithm<ref>{{Cite journal |last=Grinchtein |first=Olga |last2=Jonsson |first2=Bengt |last3=Leucker |first3=Martin |date=2010-10 |title=Learning of event-recording automata |url=https://linkinghub.elsevier.com/retrieve/pii/S0304397510003944 |journal=Theoretical Computer Science |language=en |volume=411 |issue=47 |pages=4029–4054 |doi=10.1016/j.tcs.2010.07.008}}</ref>. In essence, this algorithm is a way for programs to learn complex systems through the process of trial and error of educated guesses, to determine the behavior the system. THROUGH THE RESPONSES.... RELOCATE This a<u>lgorithm addresses the problem of identifying an unknown [[Set (mathematics)|set]]</u> through the use of a minimally adequate Teacher (MAT). The MAT provides yes or no answers to membership [[queries]], SAYING whether an input is a member of the unknown set, and equivalence queries, saying whether a description of the set is accurate or not. The Learner uses responses from the Teacher to refine its understanding of the set S in [[Time complexity|polynomial time]]<ref name=":0">{{Cite journal |last=Vaandrager |first=Frits |date=2017-01-23 |title=Model learning |url=https://dl.acm.org/doi/10.1145/2967606 |journal=Communications of the ACM |language=en |volume=60 |issue=2 |pages=86–95 |doi=10.1145/2967606 |issn=0001-0782}}</ref>. Though Angluin's paper was published in 1987, a 2017 article by computer science <u>Professor Frits Vaandrage</u>r says "the most efficient learning algorithms that are being used today all follow Angluin's approach of a minimally adequate teacher"<ref name=":0" />. |
||
=== '''Learning from Noisy Examples''' === |
=== '''Learning from Noisy Examples''' === |
||
Angluin's work on learning from noisy examples<ref>{{Cite journal |last=Angluin |first=Dana |last2=Laird |first2=Philip |date=1988-04 |title=Learning from noisy examples |url=http://link.springer.com/10.1007/BF00116829 |journal=Machine Learning |language=en |volume=2 |issue=4 |pages=343–370 |doi=10.1007/BF00116829 |issn=0885-6125}}</ref> has also been very influential to the field of [[machine learning]]. Her work addresses the problem of adapting learning algorithms to cope with incorrect training examples ([[noisy data]]). Angluin's study shows that algorithms exist for learning in the presence of noisy data in [[non-trivial]] domains and propose how the ideas might be used in more general settings. |
Angluin's work on learning from noisy examples<ref>{{Cite journal |last=Angluin |first=Dana |last2=Laird |first2=Philip |date=1988-04 |title=Learning from noisy examples |url=http://link.springer.com/10.1007/BF00116829 |journal=Machine Learning |language=en |volume=2 |issue=4 |pages=343–370 |doi=10.1007/BF00116829 |issn=0885-6125}}</ref> has also been very influential to the field of [[machine learning]]. Her work addresses the problem of adapting learning algorithms to cope with incorrect training examples ([[noisy data]]). Angluin's study shows that algorithms exist for learning in the presence of noisy data <u>in [[non-trivial]] domains</u> and propose how the ideas <u>might be used in more general settings.</u> |
||
=== '''Other Works''' === |
=== '''Other Works''' === |
||
Revision as of 18:42, 5 October 2023
Research
L* Algorithm
Angluin has written highly cited papers on computational learning theory, particularly in the context of learning regular language sets from membership and equivalence queries using the L* algorithm[1]. In essence, this algorithm is a way for programs to learn complex systems through the process of trial and error of educated guesses, to determine the behavior the system. THROUGH THE RESPONSES.... RELOCATE This algorithm addresses the problem of identifying an unknown set through the use of a minimally adequate Teacher (MAT). The MAT provides yes or no answers to membership queries, SAYING whether an input is a member of the unknown set, and equivalence queries, saying whether a description of the set is accurate or not. The Learner uses responses from the Teacher to refine its understanding of the set S in polynomial time[2]. Though Angluin's paper was published in 1987, a 2017 article by computer science Professor Frits Vaandrager says "the most efficient learning algorithms that are being used today all follow Angluin's approach of a minimally adequate teacher"[2].
Learning from Noisy Examples
Angluin's work on learning from noisy examples[3] has also been very influential to the field of machine learning. Her work addresses the problem of adapting learning algorithms to cope with incorrect training examples (noisy data). Angluin's study shows that algorithms exist for learning in the presence of noisy data in non-trivial domains and propose how the ideas might be used in more general settings.
Other Works
In distributed computing, she co-invented the population protocol model and studied the problem of consensus. In probabilistic algorithms, she has studied randomized algorithms for Hamiltonian circuits and matchings.
- ^ Grinchtein, Olga; Jonsson, Bengt; Leucker, Martin (2010-10). "Learning of event-recording automata". Theoretical Computer Science. 411 (47): 4029–4054. doi:10.1016/j.tcs.2010.07.008.
{{cite journal}}: Check date values in:|date=(help) - ^ a b Vaandrager, Frits (2017-01-23). "Model learning". Communications of the ACM. 60 (2): 86–95. doi:10.1145/2967606. ISSN 0001-0782.
- ^ Angluin, Dana; Laird, Philip (1988-04). "Learning from noisy examples". Machine Learning. 2 (4): 343–370. doi:10.1007/BF00116829. ISSN 0885-6125.
{{cite journal}}: Check date values in:|date=(help)