Articoli correlati a Learning Subsequential Transducers. A Categorical Approach
Sinossi
Master's Thesis from the year 2024 in the subject Computer Sciences - Artificial Intelligence, Università degli Studi di Milano (Dipartimento di Matematica), course: Corso di Laurea Magistrale in Matematica, language: English, abstract: In this thesis, an algorithm for learning subsequential transducers is presented from two different perspectives: as an extension of Angluin's algorithm for learning deterministic finite automata and as an instantiation of a more generic categorical algorithm valid for a larger class of automata. The adopted categorical approach considers automata as functors from a category representing words to a certain output category. Some sufficient properties for yielding the existence of minimal automata are presented, together with some additional hypotheses relative to termination to ensure the correctness of the generic algorithm. Remarkably, the conditions required in Angluin's original algorithm and in its extended version for subsequential transducers naturally arise as the generic categorical algorithm is instantiated with the proper output categories. It is not uncommon to understand facts, processes and results better by looking at them from above: learning is not an exception. Learning is a crucial area in computer science, especially in artificial intelligence: knowing how to deal with communication, mistakes and experience plays an essential role to progress in learning. But even more important is learning what makes communication possible: a language. A language can be initially thought as a subset of a set of words over an alphabet, always supposed to be finite.The relation between languages and automata has become clearer and clearer in the last decades, since Noam Chomsky gave a mathematical model of a grammar in the second half of the last century. A deterministic finite automaton accepts a language, called regular, and for every regular language there exists a deterministic finite automaton being minimal, i.e. with a minim
Le informazioni nella sezione "Riassunto" possono far riferimento a edizioni diverse di questo titolo.
Compra nuovo
Visualizza questo articolo
EUR 47,95
EUR 11,00 per la spedizione da Germania a Italia
Destinazione, tempi e costiRisultati della ricerca per Learning Subsequential Transducers. A Categorical Approach
Foto dell'editore
Learning Subsequential Transducers. A Categorical Approach
Print on DemandDa: BuchWeltWeit Ludwig Meier e.K., Bergisch Gladbach, Germania
Valutazione del venditore 5 su 5 stelle
Taschenbuch. Condizione: Neu. This item is printed on demand - it takes 3-4 days longer - Neuware 92 pp. Englisch. Codice articolo 9783389015650
Quantità: 2 disponibili
Immagini fornite dal venditore
Learning Subsequential Transducers. A Categorical Approach
Da: AHA-BUCH GmbH, Einbeck, Germania
Valutazione del venditore 5 su 5 stelle
Taschenbuch. Condizione: Neu. Druck auf Anfrage Neuware - Printed after ordering - Master's Thesis from the year 2024 in the subject Computer Sciences - Artificial Intelligence, Università degli Studi di Milano (Dipartimento di Matematica), course: Corso di Laurea Magistrale in Matematica, language: English, abstract: In this thesis, an algorithm for learning subsequential transducers is presented from two different perspectives: as an extension of Angluin's algorithm for learning deterministic finite automata and as an instantiation of a more generic categorical algorithm valid for a larger class of automata. The adopted categorical approach considers automata as functors from a category representing words to a certain output category. Some sufficient properties for yielding the existence of minimal automata are presented, together with some additional hypotheses relative to termination to ensure the correctness of the generic algorithm. Remarkably, the conditions required in Angluin¿s original algorithm and in its extended version for subsequential transducers naturally arise as the generic categorical algorithm is instantiated with the proper output categories.It is not uncommon to understand facts, processes and results better by looking at them from above: learning is not an exception. Learning is a crucial area in computer science, especially in artificial intelligence: knowing how to deal with communication, mistakes and experience plays an essential role to progress in learning. But even more important is learning what makes communication possible: a language. A language can be initially thought as a subset of a set of words over an alphabet, always supposed to be finite.The relation between languages and automata has become clearer and clearer in the last decades, since Noam Chomsky gave a mathematical model of a grammar in the second half of the last century.A deterministic finite automaton accepts a language, called regular, and for every regular language there exists a deterministic finite automaton being minimal, i.e. with a minimal number of states, that accepts it. An analogous thing happens for subsequential transducers, which are automata more complex than deterministic finite automata: in this case, some partial functions, called subsequential, from a set of words to another over possibly different alphabets are accepted;asubsequential transducer accepts a particular subsequential function and for every subsequential function there exists a minimal subsequential transducer accepting it. This is the reason why learning regular languages and subsequential functions may be pursued by learning the minimal automata accepting them. Codice articolo 9783389015650
Quantità: 1 disponibili
Immagini fornite dal venditore
Learning Subsequential Transducers. A Categorical Approach
Editore:
GRIN Verlag, GRIN Verlag Apr 2024, 2024
ISBN 10: 3389015655
ISBN 13: 9783389015650
Nuovo
Taschenbuch
Da: buchversandmimpf2000, Emtmannsberg, BAYE, Germania
Valutazione del venditore 5 su 5 stelle
Taschenbuch. Condizione: Neu. Neuware -Master's Thesis from the year 2024 in the subject Computer Sciences - Artificial Intelligence, Università degli Studi di Milano (Dipartimento di Matematica), course: Corso di Laurea Magistrale in Matematica, language: English, abstract: In this thesis, an algorithm for learning subsequential transducers is presented from two different perspectives: as an extension of Angluin's algorithm for learning deterministic finite automata and as an instantiation of a more generic categorical algorithm valid for a larger class of automata. The adopted categorical approach considers automata as functors from a category representing words to a certain output category. Some sufficient properties for yielding the existence of minimal automata are presented, together with some additional hypotheses relative to termination to ensure the correctness of the generic algorithm. Remarkably, the conditions required in Angluin¿s original algorithm and in its extended version for subsequential transducers naturally arise as the generic categorical algorithm is instantiated with the proper output categories.It is not uncommon to understand facts, processes and results better by looking at them from above: learning is not an exception. Learning is a crucial area in computer science, especially in artificial intelligence: knowing how to deal with communication, mistakes and experience plays an essential role to progress in learning. But even more important is learning what makes communication possible: a language. A language can be initially thought as a subset of a set of words over an alphabet, always supposed to be finite.The relation between languages and automata has become clearer and clearer in the last decades, since Noam Chomsky gave a mathematical model of a grammar in the second half of the last century.A deterministic finite automaton accepts a language, called regular, and for every regular language there exists a deterministic finite automaton being minimal, i.e. with a minimal number of states, that accepts it. An analogous thing happens for subsequential transducers, which are automata more complex than deterministic finite automata: in this case, some partial functions, called subsequential, from a set of words to another over possibly different alphabets are accepted;asubsequential transducer accepts a particular subsequential function and for every subsequential function there exists a minimal subsequential transducer accepting it. This is the reason why learning regular languages and subsequential functions may be pursued by learning the minimal automata accepting them. 92 pp. Englisch. Codice articolo 9783389015650
Quantità: 2 disponibili
Foto dell'editore
Learning Subsequential Transducers. A Categorical Approach
Da: California Books, Miami, FL, U.S.A.
Valutazione del venditore 5 su 5 stelle
Condizione: New. Codice articolo I-9783389015650
Quantità: Più di 20 disponibili
Immagini fornite dal venditore
Learning Subsequential Transducers. A Categorical Approach
Da: preigu, Osnabrück, Germania
Valutazione del venditore 5 su 5 stelle
Taschenbuch. Condizione: Neu. Learning Subsequential Transducers. A Categorical Approach | Riccardo Stabile | Taschenbuch | Englisch | 2024 | GRIN Verlag | EAN 9783389015650 | Verantwortliche Person für die EU: preigu, Ansas Meyer, Lengericher Landstr. 19, 49078 Osnabrück, mail[at]preigu[dot]de | Anbieter: preigu. Codice articolo 129072407
Quantità: 5 disponibili
Foto dell'editore
Learning Subsequential Transducers. A Categorical Approach
Da: dsmbooks, Liverpool, Regno Unito
Valutazione del venditore 4 su 5 stelle
paperback. Condizione: New. New. book. Codice articolo D8S0-3-M-3389015655-6
Quantità: 1 disponibili