Competing with Gaussian linear experts

Fedor Zhdanov; Vladimir Vovk

Competing with Gaussian linear experts

Research output: Working paper

166 Downloads (Pure)

Abstract

We study the problem of online regression. We prove a theoretical bound on the square loss of Ridge Regression. We do not make any assumptions about input vectors or outcomes. We also show that Bayesian Ridge Regression can be thought of as an online algorithm competing with all the Gaussian linear experts.

Original language	English
Publication status	Published - 24 Oct 2009

Keywords

cs.LG

Access to Document

pdf

PCP: Practical Competitive Prediction
Vovk, V., Gammerman, A., Kalnishkan, Y., Chernov, A. & Zhdanov, F.
Eng & Phys Sci Res Council EPSRC
10/10/07 → 22/11/10
Project: Research

Cite this

@techreport{305e4d6c2c624a87b2ddcce9aebddf18,

title = "Competing with Gaussian linear experts",

abstract = "We study the problem of online regression. We prove a theoretical bound on the square loss of Ridge Regression. We do not make any assumptions about input vectors or outcomes. We also show that Bayesian Ridge Regression can be thought of as an online algorithm competing with all the Gaussian linear experts.",

keywords = "cs.LG",

author = "Fedor Zhdanov and Vladimir Vovk",

year = "2009",

month = oct,

day = "24",

language = "English",

type = "WorkingPaper",

}

TY - UNPB

T1 - Competing with Gaussian linear experts

AU - Zhdanov, Fedor

AU - Vovk, Vladimir

PY - 2009/10/24

Y1 - 2009/10/24

N2 - We study the problem of online regression. We prove a theoretical bound on the square loss of Ridge Regression. We do not make any assumptions about input vectors or outcomes. We also show that Bayesian Ridge Regression can be thought of as an online algorithm competing with all the Gaussian linear experts.

AB - We study the problem of online regression. We prove a theoretical bound on the square loss of Ridge Regression. We do not make any assumptions about input vectors or outcomes. We also show that Bayesian Ridge Regression can be thought of as an online algorithm competing with all the Gaussian linear experts.

KW - cs.LG

M3 - Working paper

BT - Competing with Gaussian linear experts

ER -

Competing with Gaussian linear experts

Abstract

Keywords

Access to Document

Projects

PCP: Practical Competitive Prediction

Cite this