Learning in neural spatial interaction models: A statistical perspective

Fischer, Manfred M. ORCID: https://orcid.org/0000-0002-0033-2510 (2002) Learning in neural spatial interaction models: A statistical perspective. Journal of Geographical Systems, 4 (3). pp. 287-299. ISSN 1435-5949


Download (448kB)


In this paper we view learning as an unconstrained non-linear minimization problem in which the objective function is defined by the negative log-likelihood function and the search space by the parameter space of an origin constrained product unit neural spatial interaction model. We consider Alopex based global search, as opposed to local search based upon backpropagation of gradient descents, each in combination with the bootstrapping pairs approach to solve the maximum likelihood learning problem. Interregional telecommunication traffic flow data from Austria are used as test bed for comparing the performance of the two learning procedures. The study illustrates the superiority of Alopex based global search, measured in terms of Kullback and Leibler's information criterion.

Item Type: Article
Additional Information: To see the final version of this paper please visit the publisher's website. Access to the published version requires a subscription.
Keywords: Maximum likelihood learning, local search, global search, backpropagation of gradient descents, Alopex procedure, origin constrained neural spatial interaction model
Classification Codes: JEL C45, C51
Divisions: Departments > Sozioökonomie > Wirtschaftsgeographie und Geoinformatik > Fischer
Version of the Document: Accepted for Publication
Depositing User: Gertraud Novotny
Date Deposited: 04 Apr 2017 08:03
Last Modified: 04 Nov 2019 14:16
Related URLs:
FIDES Link: https://bach.wu.ac.at/d/research/results/20200/
URI: https://epub.wu.ac.at/id/eprint/5503


View Item View Item


Downloads per month over past year

View more statistics