We present a algorithm for synthesizing programs from input/output examples of their behavior. This method is a prototype of a feasible inductive inference algorithm. It is able to synthesize programs from a considerably small number of examples, which, in fact, provide only incomplete information, in general. The main computational work performed during the synthesis process consists in deducations of term equations and inequalities. The investigated synthesis algorithm is well-structured and assumes some basic knowledge formalized as a heterogeneous signature with some first order axioms. We introduce this synthesis algorithm in detail by means of a particular program for a sorting algorithm.


Inductive Inference Sorting Algorithm Synthesis Algorithm Term Equation Finite Deterministic Automaton 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. /1/.
    Angluin, D. and Smith, C.H.: Inductive inference-theory and method. Computing Surveys, Vol. 15, No.3, Sept. 1983, 237–269Google Scholar
  2. /2/.
    Knuth,D.E.: The art of computer programming, Vol.3, sorting and searching, Adison-Wesley, 1973Google Scholar
  3. /3/.
    Summers, P.D.: A methodology for LISP program construction from examples, J.A.C.M.,24, 1978, 161–175Google Scholar
  4. /4/.
    Trachtenbrot, B.A. and Barzdin, J.M.: Finite Automata-Synthesis and behaviour (russ.), Moskau 1970Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1986

Authors and Affiliations

  • Steffen Lange
    • 1
  1. 1.Department of MathematicsSteffen Lange, Humboldt University BerlinBerlin

Personalised recommendations