Learning local substitutable context-free languages from positive examples in polynomial time and data by reduction

To study more formally the approach by reduction initiated by ReGLiS, we propose a formal characterization of the grammars in reduced normal form (RNF) which can be learned by this approach. A modification of the core of ReGLiS is then proposed to ensure returning RNF grammars in polynomial time. This enables us to show that local substitutable languages represented by RNF context-free grammars are identifiable in polynomial time and thick data (IPTtD) from positive examples.

[1]  Jorge Calera-Rubio,et al.  Identifying Left-Right Deterministic Linear Languages , 2004, ICGI.

[2]  Rajeev Motwani,et al.  Introduction to automata theory, languages, and computation - international edition, 2nd Edition , 2003 .

[3]  Ryo Yoshinaka.  Learning efficiency of very simple grammars from positive data , 2009, Theor. Comput. Sci..

[4]  Alexander Clark,et al.  Learning trees from strings: a strong learning algorithm for some context-free grammars , 2013, J. Mach. Learn. Res..

[5]  José M. Sempere,et al.  A Characterization of Even Linear Languages and its Application to the Learning Problem , 1994, ICGI.

[6]  Ryo Yoshinaka,et al.  Efficiency in the Identification in the Limit Learning Paradigm , 2016 .

[7]  Ryo Yoshinaka,et al.  Learning Mildly Context-Sensitive Languages with Multidimensional Substitutability from Positive Data , 2009, ALT.

[8]  Alexander Clark,et al.  Polynomial Identification in the Limit of Substitutable Context-free Languages , 2005 .

[9]  Pieter W. Adriaans,et al.  The EMILE 4.1 Grammar Induction Toolbox , 2002, ICGI.

[10]  Yuji Takada,et al.  A Hierarchy of Language Families Learnable by Regular Language Learners , 1994, ICGI.

[11]  Colin de la Higuera.  Characteristic Sets for Polynomial Grammatical Inference , 1997 .

[12]  Ryo Yoshinaka,et al.  Distributional learning of parallel multiple context-free grammars , 2013, Machine Learning.

[13]  Jacques Nicolas,et al.  Locally Substitutable Languages for Enhanced Inductive Leaps , 2012, ICGI.

[14]  Jacques Nicolas,et al.  A bottom-up efficient algorithm learning substitutable languages from positive examples , 2014, ICGI.

[15]  Ryo Yoshinaka,et al.  Identification in the Limit of k, l-Substitutable Context-Free Languages , 2008, ICGI.

[16]  Ryo Yoshinaka,et al.  Efficient learning of multiple context-free languages with multidimensional substitutability from positive data , 2011, Theor. Comput. Sci..

[17]  Eytan Ruppin,et al.  Unsupervised learning of natural languages , 2006 .

[18]  Colin de la Higuera,et al.  Inferring Deterministic Linear Languages , 2002, COLT.

[19]  Alexander Clark,et al.  Distributional Learning of Some Context-Free Languages with a Minimally Adequate Teacher , 2010, ICGI.

[20]  Franco M. Luque,et al.  PAC-Learning Unambiguous k, l-NTS <= Languages , 2010, ICGI.

[21]  Alexander Clark,et al.  A Language Theoretic Approach to Syntactic Structure , 2011, MOL.

[22]  Zellig S. Harris,et al.  Distributional Structure , 1954 .

[23]  Menno van Zaanen.  ABL: Alignment-Based Learning , 2000, COLING.

[24]  Luc Boasson,et al.  NTS Languages Are Deterministic and Congruential , 1985, J. Comput. Syst. Sci..

[25]  Yuji Takada.  Grammatical Interface for Even Linear Languages Based on Control Sets , 1988, Inf. Process. Lett..