Filters








1,904 Hits in 5.1 sec

Recent advances of grammatical inference

Yasubumi Sakakibara
1997 Theoretical Computer Science  
In this paper, we provide a survey of recent advances in the field "Grammatical Inference" with a particular emphasis on the results concerning the learnability of target classes represented by deterministic  ...  finite automata, context-free grammars, hidden Markov models, stochastic contextfree grammars, simple recurrent neural networks, and case-based representations. * E-mail: yasu@j.dendai.ac.jp. 03043975  ...  Finally, we gratefully acknowledge the reports of anonymous referees. Their significant comments and constructive suggestions helped very much to improve the paper.  ... 
doi:10.1016/s0304-3975(97)00014-5 fatcat:4a4minst2nerti32iagjxf25ty

Learning Stochastic Finite Automata [chapter]

Colin de la Higuera, Jose Oncina
2004 Lecture Notes in Computer Science  
In the setting of identification in the limit with probability one, we prove that stochastic deterministic finite automata cannot be identified from only a polynomial quantity of data.  ...  Stochastic deterministic finite automata have been introduced and are used in a variety of settings. We report here a number of results concerning the learnability of these finite state machines.  ...  Acknowledgement: The authors thank Franck Thollard for pointing out to them the result from section 4.  ... 
doi:10.1007/978-3-540-30195-0_16 fatcat:4mlmcctfsvcnnbqnso5v4b6w34

Page 1554 of Mathematical Reviews Vol. 57, Issue 4 [page]

1979 Mathematical Reviews  
Starke, Multitape automata and languages (pp. 13-26); Hans-Dieter Burkhard, On theory of state- identification experiments at finite non-deterministic automata (pp. 27-40); Jan Kolenitka, A comparison  ...  M. 57 #11973 On the question of the construction of automata that have a limited capability to detect faults occurring in them. (Russian.  ... 

A bibliographical study of grammatical inference

Colin de la Higuera
2005 Pattern Recognition  
There is no uniform literature on the subject and one can find many papers with original definitions or points of view.  ...  Some of these papers are essential and should constitute a common background to research in the area, whereas others are specialized on particular problems or techniques, but can be of great help on specific  ...  Description of the process of identification in the limit with probability 1, and useful tools can be found in Angluin's unpublished report [51] .  ... 
doi:10.1016/j.patcog.2005.01.003 fatcat:62qwskiqcvddjobakbdshwebqq

Quantum inductive inference by finite automata

R. Freivalds, R.F. Bonner
2008 Theoretical Computer Science  
Smith Memory limited inductive inference machines, Springer Lecture Notes in Computer Science 621 (1992) 19-29] proved that probabilistic limited memory inductive inference machines can learn with probability  ...  We introduce quantum limited memory inductive inference machines as quantum finite automata acting as inductive inference machines.  ...  Acknowledgment The first author's research was supported by Grant No. 05.1528 from the Latvian Council of Science, Contract IST-1999-11234 (QAIP) from the European Commission, and the Swedish Institute  ... 
doi:10.1016/j.tcs.2008.02.023 fatcat:4vvokuge2fb3blvri2clio77za

Ten Open Problems in Grammatical Inference [chapter]

Colin de la Higuera
2006 Lecture Notes in Computer Science  
We propose 10 different open problems in the field of grammatical inference. In all cases, problems are theoretically oriented but correspond to practical questions.  ...  They cover the areas of polynomial learning models, learning from ordered alphabets, learning deterministic Pomdps, learning negotiation processes, learning from context-free background knowledge.  ...  Work with Henning Fernau on polynomial learning is where the ideas in section 3 come from. Philippe Jaillon gave me the initial ideas for the negotiation problem in section 9. Discussions with  ... 
doi:10.1007/11872436_4 fatcat:5snm4lpumbhw5e7ffh5vwxzhum

Learning stochastic finite automata from experts [chapter]

Colin de la Higuera
1998 Lecture Notes in Computer Science  
We present in this paper a new learning problem called learning distributions from experts. In the case we study the experts are stochastic deterministic finite automata (sdfa).  ...  In this paper we prove that although a polynomial identification with probability one is not always possible, a wide class of automata can successfully, and for this criterion, be identified.  ...  Perhaps even a second expert can help : one can use the amount of times experts agree on the ordering of specific strings as a compatibility test.  ... 
doi:10.1007/bfb0054066 fatcat:mtep6bidkbghbkvnkchqtym6q4

Approximate identification of automata

B.R. Gaines
1975 Electronics Letters  
A technique is described for the identification of probabilistic and other non-deterministic automata from sequences of their input/output behaviour.  ...  For a given number of states the models obtained are optimal in well defined senses, one related to least-mean-square approximation and the other to Shannon entropy.  ...  Practically, its use as a system identification technique is feasible in simple cases, but is limited by computation time to models with up to 10 states.  ... 
doi:10.1049/el:19750342 fatcat:3jun2kpgf5gc7m5j3nmwe5drxi

Page 1079 of Automation and Remote Control Vol. 37, Issue 7 [page]

1976 Automation and Remote Control  
LARGE STOCHASTIC SYSTEMS OF FINITE AUTOMATA AND NEURAL NETS, I A. R.  ...  With this, even if the orig- inal system is a deterministic one, the massive character of the interactions within it, as a mule, compel one to consider the initial data, and the parameters occurring in  ... 

Learning Stochastic Finite Automata for Musical Style Recognition [chapter]

Colin de la Higuera, Frédéric Piat, Frédéric Tantini
2006 Lecture Notes in Computer Science  
Stochastic deterministic finite automata have been introduced and are used in a variety of settings.  ...  Through grammatical inference these automata are learned and new pieces of music can be parsed. We show that this works by proposing promising classification results and discuss further work.  ...  Acknowledgement: The authors are grateful to Pedro Cruz for his benchmarks and for many ideas used in this work. They also thank Thierry Murgue and Franck Thollard for help with Mdi and parsers.  ... 
doi:10.1007/11605157_31 fatcat:k3m66q4ndjfshmcktmk7kougua

Probabilistic finite-state machines - part II

E. Vidal, F. Thollard, C. de la Higuera, F. Casacuberta, R.C. Carrasco
2005 IEEE Transactions on Pattern Analysis and Machine Intelligence  
In this Part II, we study the relations between probabilistic finite-state automata and other well-known devices that generate strings like hidden Markov models and n-grams and provide theorems, algorithms  ...  Probabilistic finite-state machines are used today in a variety of areas in pattern recognition or in fields to which pattern recognition is linked.  ...  This work has been partially supported by the Spanish project TIC2003-08681-C02 and the IST Programme of the European Community, under the PASCAL Network of Excellence, IST-2002-506778.  ... 
doi:10.1109/tpami.2005.148 pmid:16013757 fatcat:vaoopt4ypzffzpv53pxx2hodpy

Learning Languages with Help [chapter]

Christopher Kermorvant, Colin de la Higuera
2002 Lecture Notes in Computer Science  
We propose a general setting to deal with these cases and provide algorithms that can learn deterministic finite automata in these conditions.  ...  Furthermore the number of examples needed to correctly identify diminishes drastically with the quality of the added information.  ...  Polynomial Identification in the Limit from Given Data The question of learning grammars or automata with help can be tackled in a variety of learning models.  ... 
doi:10.1007/3-540-45790-9_13 fatcat:adbjrahojfejxieqmglom6vnay

Learning deterministic regular grammars from stochastic samples in polynomial time

Rafael C. Carrasco, Jose Oncina
1999 RAIRO - Theoretical Informatics and Applications  
In this paper, the identification of stochastic regular languages is addressed.  ...  For this purpose, we propose a class of algorithms which allow for the identification of the structure of the minimal stochastic automaton generating the language.  ...  Identification in the limit means that only finitely many changes of hypothesis take place before a correct one is found.  ... 
doi:10.1051/ita:1999102 fatcat:rw2vcb2qtnfo7ma5cns3dffuum

Page 1036 of Mathematical Reviews Vol. , Issue 93b [page]

1993 Mathematical Reviews  
The general result: The complexity of stochastic automata needed to recognize a given language is usually below the complexity of deterministic automata needed for the same problem.  ...  -IITKH-C) On characterization of cellular automata with matrix algebra.  ... 

On the Complexity of Causal Models

B. R. Gaines
1976 IEEE Transactions on Systems, Man and Cybernetics  
This correspondence provides an automata-theoretic explanation of this phenomenon by analyzing the performance of an optimal modeler observing the behaviour of a system and forming a minimal state model  ...  of it.  ...  ACKNOWLEDGEMENT I am grateful to Ian Witten of this Department for his critical comments on the first draft of this correspondence, and to Dr.  ... 
doi:10.1109/tsmc.1976.5408397 fatcat:gepzthwn5vcx5m7httjiyabake
« Previous Showing results 1 — 15 out of 1,904 results