Computable Learning, Neural Networks and Institutions

Abstract

After showing the positive role that an institution can play in the learning process, we motivate the intuition for why-in the context of a major change in the environment-the more rigid and strictly specialized an institution is, the longer and more complex the learning process will be of any actor subject to the by-now obsolete institution. In particular, the inductive adaptation of economic actors (firms in our setting) to a new environment (such as the one caused by transition to a market economy) is slowed by the very existence of some institution or organizational setting that had emerged in the original environment as a superior tool of induction.

The author would like to thank Professors K. Velupillai, A. Leijonhufvud, and L. Punzo for their valuable comments on various partial versions of this work. Professor G. Mondello has given strong encouragement and Dr S. Bartolini endured long precious discussions. None of them is responsible for any error contained in this paper which was presented for the first time at the conference on Computing in Economics and Finance, Geneva 1996. A Human Capital and Mobility grant is gratefully acknowledged.