Variable metric stochastic approximation theory

Peter Sunehag; Jochen Trumpf; S. V.N. Vishwanathan; Nicol N. Schraudolph

Variable metric stochastic approximation theory

Peter Sunehag^*, Jochen Trumpf, S. V.N. Vishwanathan, Nicol N. Schraudolph

^*Corresponding author for this work

Research output: Contribution to journal › Conference article › peer-review

10 Citations (Scopus)

Abstract

We provide a variable metric stochastic approximation theory. In doing so, we provide a convergence theory for a large class of online variable metric methods including the recently introduced online versions of the BFGS algorithm and its limited-memory LBFGS variant. We also discuss the implications of our results for learning from expert advice.

Original language	English
Pages (from-to)	560-566
Number of pages	7
Journal	Journal of Machine Learning Research
Volume	5
Publication status	Published - 2009
Event	12th International Conference on Artificial Intelligence and Statistics, AISTATS 2009 - Clearwater, FL, United States Duration: 16 Apr 2009 → 18 Apr 2009

Cite this

@article{722ff2e54b2845d481d1ec5c3825ec26,

title = "Variable metric stochastic approximation theory",

abstract = "We provide a variable metric stochastic approximation theory. In doing so, we provide a convergence theory for a large class of online variable metric methods including the recently introduced online versions of the BFGS algorithm and its limited-memory LBFGS variant. We also discuss the implications of our results for learning from expert advice.",

author = "Peter Sunehag and Jochen Trumpf and Vishwanathan, {S. V.N.} and Schraudolph, {Nicol N.}",

year = "2009",

language = "English",

volume = "5",

pages = "560--566",

journal = "Journal of Machine Learning Research",

issn = "1532-4435",

publisher = "Microtome Publishing",

note = "12th International Conference on Artificial Intelligence and Statistics, AISTATS 2009 ; Conference date: 16-04-2009 Through 18-04-2009",

}

TY - JOUR

T1 - Variable metric stochastic approximation theory

AU - Sunehag, Peter

AU - Trumpf, Jochen

AU - Vishwanathan, S. V.N.

AU - Schraudolph, Nicol N.

PY - 2009

Y1 - 2009

N2 - We provide a variable metric stochastic approximation theory. In doing so, we provide a convergence theory for a large class of online variable metric methods including the recently introduced online versions of the BFGS algorithm and its limited-memory LBFGS variant. We also discuss the implications of our results for learning from expert advice.

AB - We provide a variable metric stochastic approximation theory. In doing so, we provide a convergence theory for a large class of online variable metric methods including the recently introduced online versions of the BFGS algorithm and its limited-memory LBFGS variant. We also discuss the implications of our results for learning from expert advice.

UR - http://www.scopus.com/inward/record.url?scp=84862282880&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:84862282880

SN - 1532-4435

VL - 5

SP - 560

EP - 566

JO - Journal of Machine Learning Research

JF - Journal of Machine Learning Research

T2 - 12th International Conference on Artificial Intelligence and Statistics, AISTATS 2009

Y2 - 16 April 2009 through 18 April 2009

ER -

Variable metric stochastic approximation theory

Abstract

Other files and links

Fingerprint

Cite this