Abstract
We carefully investigate the online version of PCA, where in each trial a learning algorithm plays a k-dimensional subspace, and suffers the compression loss on the next instance when projected into the chosen subspace. In this setting, we give regret bounds for two popular online algorithms, Gradient Descent (GD) and Matrix Exponentiated Gradient (MEG). We show that both algorithms are essentially optimal in the worst-case when the regret is expressed as a function of the number of trials. This comes as a surprise, since MEG is commonly believed to perform sub-optimally when the instances are sparse. This different behavior of MEG for PCA is mainly related to the non-negativity of the loss in this case, which makes the PCA setting qualitatively different from other settings studied in the literature. Furthermore, we show that when considering regret bounds as a function of a loss budget, MEG remains optimal and strictly outperforms GD.
Next, we study a generalization of the online PCA problem, in which the Nature is allowed to play with dense instances, which are positive matrices with bounded largest eigenvalue. Again we can show that MEG is optimal and strictly better than GD in this setting.
This research was supported by the NSF grant IIS-0917397, while the second author was visiting UC Santa Cruz.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Abernethy, J., Agarwal, A., Bartlett, P.L., Rakhlin, A.: A stochastic view of optimal regret through minimax duality. In: COLT (2009)
Abernethy, J., Warmuth, M.K., Yellin, J.: When random play is optimal against an adversary. In: COLT, pp. 437–446 (2008)
Azoury, K.S., Warmuth, M.K.: Relative loss bounds for on-line density estimation with the exponential family of distributions. Machine Learning 43(3), 211–246 (2001)
Cesa-Bianchi, N., Freund, Y., Haussler, D., Helmbold, D.P., Schapire, R.E., Warmuth, M.K.: How to use expert advice. J. ACM 44(3), 427–485 (1997)
Cesa-Bianchi, N., Long, P.M., Warmuth, M.K.: Worst-case quadratic loss bounds for prediction using linear functions and gradient descent. IEEE Trans. Neural Netw. Learning Syst. 7(3), 604–619 (1996)
Cesa-Bianchi, N., Long, P.M., Warmuth, M.K.: Worst-case quadratic loss bounds for prediction using linear functions and gradient descent. IEEE Trans. Neural Netw. Learning Syst. 7(3), 604–619 (1996)
Cesa-Bianchi, N., Lugosi, G.: Prediction, learning, and games. Cambridge University Press (2006)
Helmbold, D.P., Warmuth, M.K.: Learning permutations with exponential weights. Journal of Machine Learning Research 10, 1705–1736 (2009)
Herbster, M., Warmuth, M.K.: Tracking the best linear predictor. Journal of Machine Learning Research 1, 281–309 (2001)
Kalai, A.T., Vempala, S.: Efficient algorithms for online decision problems. J. Comput. Syst. Sci. 71(3), 291–307 (2005)
Kivinen, J., Warmuth, M.K.: Exponentiated gradient versus gradient descent for linear predictors. Inf. Comput. 132(1), 1–63 (1997)
Koolen, W.M., Warmuth, M.K., Kivinen, J.: Hedging structured concepts. In: COLT, pp. 93–105 (2010)
Kuzmin, D., Warmuth, M.K.: Online kernel PCA with entropic matrix updates. In: ICML, pp. 465–472 (2007)
Srebro, N., Sridharan, K., Tewari, A.: On the universality of online mirror descent. In: NIPS, pp. 2645–2653 (2011)
Sridharan, K., Tewari, A.: Convex games in banach spaces. In: Proceedings of the 23nd Annual Conference on Learning Theory (COLT) (2010)
Tsuda, K., Rätsch, G., Warmuth, M.K.: Matrix exponentiated gradient updates for on-line learning and Bregman projections. Journal of Machine Learning Research 6, 995–1018 (2005)
Warmuth, M.K., Kuzmin, D.: Randomized online PCA algorithms with regret bounds that are logarithmic in the dimension. Journal of Machine Learning Research 9, 2287–2320 (2008)
Warmuth, M.K., Liao, J., Rätsch, G.: Totally corrective boosting algorithms that maximize the margin. In: ICML, pp. 1001–1008 (2006)
Warmuth, M.K., Vishwanathan, S.V.N.: Leaving the span. In: Auer, P., Meir, R. (eds.) COLT 2005. LNCS (LNAI), vol. 3559, pp. 366–381. Springer, Heidelberg (2005)
Zinkevich, M.: Online convex programming and generalized infinitesimal gradient ascent. In: Fawcett, T., Mishra, N. (eds.) ICML, pp. 928–936. AAAI Press (2003)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Nie, J., Kotłowski, W., Warmuth, M.K. (2013). Online PCA with Optimal Regrets. In: Jain, S., Munos, R., Stephan, F., Zeugmann, T. (eds) Algorithmic Learning Theory. ALT 2013. Lecture Notes in Computer Science(), vol 8139. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-40935-6_8
Download citation
DOI: https://doi.org/10.1007/978-3-642-40935-6_8
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-40934-9
Online ISBN: 978-3-642-40935-6
eBook Packages: Computer ScienceComputer Science (R0)