Formal Language Learning with Correction Queries

In the field of grammatical inference most of the
existing learning models have been inspired by the
process of human language acquisition. For example,
the query learning model is a representation of our
intuition that children acquire their native language
while interacting with their environment.
Apart from the well-established membership and
equivalence queries, many other types of queries have
been introduced. Nevertheless, none of these reflect
an important aspect of the way children learn
languages, namely that although they are not
explicitly provided negative examples (i.e., words
that are not in the language or ungrammatical
sentences), adults correct them when they make
mistakes. In this book we thoroughly investigate a
recently introduced type of query, called CORRECTION
QUERY, which copes with this particularity of
children language acquisition. Here are some of the
topics addressed: What is the power of correction
queries? Can we compare them with other query
learning or Gold-style learning models? Which of them
provides the learner with more information? Can we
build efficient algorithms that learn well-known
classes of languages with correction queries?

Autorentext

Cristina Tîrnauca was born in Bacau, Romania on July the 2nd

1979. In 2003 she got her BS in mathematics and computer science
      at the University of Bucharest and her BEc at the Academy of
      Economic Studies, Bucharest. In 2009 she obtains the PhD degree
      in computer science and artificial intelligence at Rovira i
      Virgili University, Tarragona, Spain.

      Klappentext

      In the field of grammatical inference most of the
      existing learning models have been inspired by the
      process of human language acquisition. For example,
      the query learning model is a representation of our
      intuition that children acquire their native language
      while interacting with their environment.
      Apart from the well-established membership and
      equivalence queries, many other types of queries have
      been introduced. Nevertheless, none of these reflect
      an important aspect of the way children learn
      languages, namely that although they are not
      explicitly provided negative examples (i.e., words
      that are not in the language or ungrammatical
      sentences), adults correct them when they make
      mistakes. In this book we thoroughly investigate a
      recently introduced type of query, called CORRECTION
      QUERY, which copes with this particularity of
      children language acquisition. Here are some of the
      topics addressed: What is the power of correction
      queries? Can we compare them with other query
      learning or Gold-style learning models? Which of them
      provides the learner with more information? Can we
      build efficient algorithms that learn well-known
      classes of languages with correction queries?