Variable Metric Stochastic Approximation Theory

Peter Sunehag, Jochen Trumpf, S.V.N. Vishwanathan, Nicol Schraudolph
; Proceedings of the Twelth International Conference on Artificial Intelligence and Statistics, PMLR 5:560-566, 2009.

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 in the areas of elicitation of properties of distributions using prediction markets and in learning from expert advice.

Cite this Paper


BibTeX
@InProceedings{pmlr-v5-sunehag09a, title = {Variable Metric Stochastic Approximation Theory}, author = {Peter Sunehag and Jochen Trumpf and S.V.N. Vishwanathan and Nicol Schraudolph}, booktitle = {Proceedings of the Twelth International Conference on Artificial Intelligence and Statistics}, pages = {560--566}, year = {2009}, editor = {David van Dyk and Max Welling}, volume = {5}, series = {Proceedings of Machine Learning Research}, address = {Hilton Clearwater Beach Resort, Clearwater Beach, Florida USA}, month = {16--18 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v5/sunehag09a/sunehag09a.pdf}, url = {http://proceedings.mlr.press/v5/sunehag09a.html}, 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 in the areas of elicitation of properties of distributions using prediction markets and in learning from expert advice.} }
Endnote
%0 Conference Paper %T Variable Metric Stochastic Approximation Theory %A Peter Sunehag %A Jochen Trumpf %A S.V.N. Vishwanathan %A Nicol Schraudolph %B Proceedings of the Twelth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2009 %E David van Dyk %E Max Welling %F pmlr-v5-sunehag09a %I PMLR %J Proceedings of Machine Learning Research %P 560--566 %U http://proceedings.mlr.press %V 5 %W PMLR %X 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 in the areas of elicitation of properties of distributions using prediction markets and in learning from expert advice.
RIS
TY - CPAPER TI - Variable Metric Stochastic Approximation Theory AU - Peter Sunehag AU - Jochen Trumpf AU - S.V.N. Vishwanathan AU - Nicol Schraudolph BT - Proceedings of the Twelth International Conference on Artificial Intelligence and Statistics PY - 2009/04/15 DA - 2009/04/15 ED - David van Dyk ED - Max Welling ID - pmlr-v5-sunehag09a PB - PMLR SP - 560 DP - PMLR EP - 566 L1 - http://proceedings.mlr.press/v5/sunehag09a/sunehag09a.pdf UR - http://proceedings.mlr.press/v5/sunehag09a.html 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 in the areas of elicitation of properties of distributions using prediction markets and in learning from expert advice. ER -
APA
Sunehag, P., Trumpf, J., Vishwanathan, S. & Schraudolph, N.. (2009). Variable Metric Stochastic Approximation Theory. Proceedings of the Twelth International Conference on Artificial Intelligence and Statistics, in PMLR 5:560-566

Related Material