Generalized Stochastic Gradient Learning

George William Evans (), Seppo Mikko Sakari Honkapohja () and Noah Williams ()

No 317, NBER Technical Working Papers from National Bureau of Economic Research, Inc

Abstract: We study the properties of generalized stochastic gradient (GSG) learning in forward-looking models. We examine how the conditions for stability of standard stochastic gradient (SG) learning both differ from and are related to E-stability, which governs stability under least squares learning. SG algorithms are sensitive to units of measurement and we show that there is a transformation of variables for which E-stability governs SG stability. GSG algorithms with constant gain have a deeper justification in terms of parameter drift, robustness and risk sensitivity.

JEL-codes: C62 C65 D83 E10 E17 (search for similar items in EconPapers)
New Economics Papers: this item is included in nep-cmp, nep-evo and nep-mac
Date: 2005-10
Note: TWP
View list of references View citations in EconPapers

Downloads: (external link)
http://www.nber.org/papers/t0317.pdf (application/pdf)
Access to the full text is generally limited to series subscribers, however if the top level domain of the client browser is in a developing country or transition economy free access is provided. More information about subscriptions and free access is available at http://www.nber.org/wwphelp.html.

Related works:
Working Paper: Generalized Stochastic Gradient Learning (2005)
Working Paper: Generalized Stochastic Gradient Learning (2005)
Working Paper: Generalized Stochastic Gradient Learning (2007)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Ordering information: This working paper can be ordered from
http://www.nber.org/papers/t0317
The price is Paper copy available by mail.

Access Statistics for this paper

More papers in NBER Technical Working Papers from National Bureau of Economic Research, Inc
Address: National Bureau of Economic Research, 1050 Massachusetts Avenue Cambridge, MA 02138, U.S.A.
Contact information at EDIRC.
Series data maintained by ().