Years and Authors of Summarized Original Work
2003; Zinkevich
Problem Definition
Suppose we are going to invest in a stock market. Our neighbor, for mysterious reasons, happens to know how the market evolves. But he cannot change his portfolio (proportions of holding stocks) once committed (to avoid being caught by regulators, say). On the other hand, we, the normal investor, do not have any inside information but can sell and buy at will. If we and our prescient neighbor invest the same amount of money, is there a (computationally feasible) way for us to perform comparably well to our neighbor, without knowing his investing strategy? Surprisingly (as contrary to our real-life experience perhaps), the answer is yes, and we will see it through the lens of online learning. Disclaimer: The reader is at his own risk if he decides to practice the beautiful theoretical results we describe below.
The online learning problem is best described as a multi-round two-person game between the...
This is a preview of subscription content, log in via an institution to check access.
Recommended Reading
Bartlett PL, Hazan E, Rakhlin A (2007) Adaptive online gradient descent. In: Platt JC, Koller D, Singer Y, Roweis ST (eds) Advances in neural information processing systems 20 (NIPS). Curran Associates, Inc., Vancouver, pp 257–269
Bubeck S, Cesa-Bianchi N (2012) Regret analysis of stochastic and nonstochastic multi-armed bandit problems. Found Trends Mach Learn 5(1):1–122
Cesa-Bianchi N, Lugosi G (2006) Prediction, learning, and games. Cambridge University Press, New York
Duchi JC, Shalev-Shwartz S, Singer Y, Tewari A (2010) Composite objective mirror descent. In: Kalai AT, Mohri M (eds) The 23rd conference on learning theory (COLT). Haifa, pp 14–26
Hazan E, Kale S (2012) Projection-free online learning. In: Langford J, Pineau J (eds) The 29th international conference on machine learning (ICML), Edinburgh. Omnipress, pp 521–528
Hazan E, Agarwal A, Kale S (2007) Logarithmic regret algorithms for online convex optimization. Mach Learn 69:169–192
Nemirovski A, Juditsky A, Lan G, Shapiro A (2009) Robust stochastic approximation approach to stochastic programming. SIAM J Optim 19(4):1574–1609
Shalev-Shwartz S (2011) Online learning and online convex optimization. Found Trends Mach Learn 4(2):107–194
Zinkevich M (2003) Online convex programming and generalized infinitesimal gradient approach. In: Fawcett T, Mishra N (eds) The 20th international conference on machine learning (ICML), Washington. AAAI Press, pp 928–936
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer Science+Business Media New York
About this entry
Cite this entry
Yu, Y. (2015). Online Learning and Optimization. In: Kao, MY. (eds) Encyclopedia of Algorithms. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-27848-8_265-2
Download citation
DOI: https://doi.org/10.1007/978-3-642-27848-8_265-2
Received:
Accepted:
Published:
Publisher Name: Springer, Berlin, Heidelberg
Online ISBN: 978-3-642-27848-8
eBook Packages: Springer Reference Computer SciencesReference Module Computer Science and Engineering