Reference Work Entry

Encyclopedia of Algorithms

pp 425-428

Learning Automata

2000; Beimel, Bergadano, Bshouty, Kushilevitz, Varricchio
  • Amos BeimelAffiliated withBen-Gurion University
  • , Francesco BergadanoAffiliated withUniversity of Torino
  • , Nader H. BshoutyAffiliated withTechnion
  • , Eyal KushilevitzAffiliated withTechnion
  • , Stefano VarricchioAffiliated withDepartment of Computer Science, University of Roma

Keywords and Synonyms

Computational learning; Machine learning; Multiplicity automata; Formal series; Boolean formulas; Multivariate polynomials

Problem Definition

This problem is concerned with the learnability of multiplicity automata in Angluin's exact learning model and applications to the learnability of functions represented by small multiplicity automata.

The Learning Model

It is the exact learning model [2]: Let f be a target function. A learning algorithm may propose to an oracle, in each step, two kinds of queries: membership queries (MQ) and equivalence queries (EQ). In a MQ it may query for the value of the function f on a particular assignment z. The response to such a query is the value f(z).1 In a EQ it may propose to the oracle a hypothesis function h. If h is equivalent to f on all input assignments then the answer to the query is YES a ...

This is an excerpt from the content