A probabilistic interpretation of the constant gain algorithm

Michele Berardi

A probabilistic interpretation of the constant gain algorithm

Economics

Research output: Preprint/Working paper › Working paper

Abstract

This paper proposes a novel interpretation of the constant gain learning algorithm through a probabilistic setting with Bayesian updating. Such framework allows to understand the gain coefficient in terms of the probability of changes in the estimated quantity.

Original language	English
Publisher	University Library of Munich
Publication status	Published - 2019

Publication series

Name	MPRA
Publisher	University Library of Munich, Germany
No.	94023

Access to Document

https://ideas.repec.org/p/pra/mprapa/94023.htmlLicence: Unspecified

Cite this

@techreport{bb92c4d0093a43f1a931a1cb4551f7a8,

title = "A probabilistic interpretation of the constant gain algorithm",

abstract = "This paper proposes a novel interpretation of the constant gain learning algorithm through a probabilistic setting with Bayesian updating. Such framework allows to understand the gain coefficient in terms of the probability of changes in the estimated quantity.",

author = "Michele Berardi",

year = "2019",

language = "English",

series = "MPRA",

publisher = "University Library of Munich",

number = "94023",

address = "Germany",

type = "WorkingPaper",

institution = "University Library of Munich",

}

TY - UNPB

T1 - A probabilistic interpretation of the constant gain algorithm

AU - Berardi, Michele

PY - 2019

Y1 - 2019

N2 - This paper proposes a novel interpretation of the constant gain learning algorithm through a probabilistic setting with Bayesian updating. Such framework allows to understand the gain coefficient in terms of the probability of changes in the estimated quantity.

AB - This paper proposes a novel interpretation of the constant gain learning algorithm through a probabilistic setting with Bayesian updating. Such framework allows to understand the gain coefficient in terms of the probability of changes in the estimated quantity.

M3 - Working paper

T3 - MPRA

BT - A probabilistic interpretation of the constant gain algorithm

PB - University Library of Munich

ER -

A probabilistic interpretation of the constant gain algorithm

Abstract

Publication series

Access to Document

Fingerprint

Cite this