Learning from Errors

We present a model of learning in which agents learn from errors. If an action turns out to be an error, the agent rejects not only that action but also neighboring actions. We find that, keepng memory of his errors, under mild assumptions an acceptable solution is asymptotically reached. Moreover, one can take advantage of big errors for a faster learning.

MoreLess

Year of publication:	2003-01-21
Authors:	Martinez-Legaz, Juan Enrique ; Soubeyran, Antoine
Institutions:	Departament d'Economia i Història Econòmica, Universitat Autònoma de Barcelona
Subject:	Learning \| errors \| fixed points \| normal form games \| best reply functions

More details

Extent:	application/pdf
Series:	UFAE and IAE Working Papers.
Type of publication:	Book / Working Paper
Language:	Czech
Notes:	2 pages long
Classification:	D83 - Search, Learning, Information and Knowledge ; C72 - Noncooperative Games
Source:	RePEc - Research Papers in Economics

Persistent link: https://ebvufind01.dmz1.zbw.eu/10005572182