Random-Walk Perturbations for Online Combinatorial Optimization

IEEE Transactions on Information Theory

Volumn 61, Issue 7, 2015, Pages 4099-4106

  Devroye, Luc ^a   Lugosi, Gábor ^b   Neu, Gergely ^c  

^a MCGILL UNIVERSITY   (Canada)

^b UNIVERSITAT POMPEU FABRA   (Spain)

^c INRIA   (France)

Author keywords

Follow the Perturbed Leader; Online combinatorial optimization; Online learning; Random walk

Indexed keywords

ALGORITHMS; COMBINATORIAL OPTIMIZATION; COMBINATORIAL SWITCHING; OPTIMIZATION; RANDOM PROCESSES; SOCIAL NETWORKING (ONLINE);

COMBINATORIAL OPTIMIZATION PROBLEMS; FOLLOW THE PERTURBED LEADER; ONLINE LEARNING; OPTIMAL ORDERING; PREDICTION WITH EXPERT ADVICE; QUANTITIES OF INTERESTS; RANDOM WALK; SWITCHING COSTS;

PROBLEM SOLVING;

EID: 84933074141     PISSN: 00189448     EISSN: None     Source Type: Journal    
DOI: 10.1109/TIT.2015.2428253     Document Type: Article

References (28)

1. L. Devroye, G. Lugosi, and G. Neu, "Prediction by random-walk perturbation," in Proc. 25th Annu. Conf. Learn. Theory (COLT), 2013, pp. 460-473.
2. N. Cesa-Bianchi and G. Lugosi, Prediction, Learning, and Games. New York, NY, USA: Cambridge Univ. Press, 2006.
3. C. Gentile and M. K. Warmuth, "Linear hinge loss and average margin," in Proc. Adv. Neural Inf. Process. Syst. (NIPS), 1998, pp. 225-231.
4. J. Kivinen and M. K. Warmuth, "Relative loss bounds for multidimensional regression problems," Mach. Learn., vol. 45, no. 3, pp. 301-329, 2001.
5. A. J. Grove, N. Littlestone, and D. Schuurmans, "General convergence results for linear discriminant updates," Mach. Learn., vol. 43, no. 3, pp. 173-210, 2001.
6. E. Takimoto and M. K. Warmuth, "Path kernels and multiplicative updates," J. Mach. Learn. Res., vol. 4, pp. 773-818, Oct. 2003.
7. A. Kalai and S. Vempala, "Efficient algorithms for online decision problems," J. Comput. Syst. Sci., vol. 71, no. 3, pp. 291-307, 2005.
8. M. K. Warmuth and D. Kuzmin, "Randomized online PCA algorithms with regret bounds that are logarithmic in the dimension," J. Mach. Learn. Res., vol. 9, no. 10, pp. 2287-2320, 2008.
9. D. P. Helmbold and M. K. Warmuth, "Learning permutations with exponential weights," J. Mach. Learn. Res., vol. 10, pp. 1705-1736, Jul. 2009.
10. E. Hazan, S. Kale, and M. K. Warmuth, "Learning rotations with little regret," in Proc. 23rd Annu. Conf. Learn. Theory (COLT), 2010, pp. 144-154.
11. W. M. Koolen, M. K. Warmuth, and J. Kivinen, "Hedging structured concepts," in Proc. 23rd Annu. Conf. Learn. Theory (COLT), 2010, pp. 93-105.
12. N. Cesa-Bianchi and G. Lugosi, "Combinatorial bandits," J. Comput. Syst. Sci., vol. 78, no. 5, pp. 1404-1422, 2012.
13. J.-Y. Audibert, S. Bubeck, and G. Lugosi, "Regret in online combinatorial optimization," Math. Oper. Res., vol. 39, no. 1, pp. 31-45, 2014.
14. N. Littlestone and M. K. Warmuth, "The weighted majority algorithm," Inf. Comput., vol. 108, no. 2, pp. 212-261, 1994.
15. V. G. Vovk, "Aggregating strategies," in Proc. 3rd Annu. Workshop Comput. Learn. Theory (COLT), 1990, pp. 371-386.
16. Y. Freund and R. E. Schapire, "A decision-theoretic generalization of on-line learning and an application to boosting," J. Comput. Syst. Sci., vol. 55, no. 1, pp. 119-139, 1997.
17. J. Hannan, "Approximation to Bayes risk in repeated play," Ann. Math. Studies, Contrib. Theory Games, , vol. 3, no. 39, pp. 97-139, 1957.
18. M. Hutter and J. Poland, "Prediction with expert advice by following the perturbed leader for general weights," in Proc. 23rd Int. Conf. Algorithmic Learn. Theory (ALT), 2004, pp. 279-293.
19. J. Poland, "FPL analysis for adaptive bandits," in Proc. 3rd Int. Symp. Stochastic Algorithms, Found. Appl. (SAGA), 2005, pp. 58-69.
20. G. Neu and G. Bartók, "An efficient algorithm for learning with semibandit feedback," in Proc. 24th Int. Conf. Algorithmic Learn. Theory (ALT), 2013, pp. 234-248.
21. D. Suehiro, K. Hatano, S. Kijima, E. Takimoto, and K. Nagano, "Online prediction under submodular constraints," in Proc. 23rd Int. Conf. Algorithmic Learn. Theory (ALT), vol. 7568. 2012, pp. 260-274.
22. S. Geulen, B. Vöcking, and M. Winkler, "Regret minimization for online buffering problems using the weighted majority algorithm," in Proc. 23rd Annu. Conf. Learn. Theory (COLT), 2010, pp. 132-143.
23. A. György and G. Neu, "Near-optimal rates for limited-delay universal lossy source coding," in Proc. IEEE Int. Symp. Inf. Theory (ISIT), Jul./Aug. 2011, pp. 2218-2222.
24. E. Even-Dar, S. M. Kakade, and Y. Mansour, "Online Markov decision processes," Math. Oper. Res., vol. 34, no. 3, pp. 726-736, 2009.
25. G. Neu, A. György, C. Szepesvári, and A. Antos, "Online Markov decision processes under bandit feedback," in Proc. Adv. Neural Inf. Process. Syst. (NIPS), vol. 23. 2011, pp. 1804-1812.
26. S. Rakhlin, O. Shamir, and K. Sridharan, "Relax and randomize: From value to algorithms," in Proc. Adv. Neural Inf. Process. Syst. (NIPS), vol. 25. 2012, pp. 2150-2158.
27. W. Feller, An Introduction to Probability Theory and Its Applications, vol. 1. New York, NY, USA: Wiley, 1968.
28. S. Boucheron, G. Lugosi, and P. Massart, Concentration Inequalities: A Nonasymptotic Theory of Independence. London, U.K.: Oxford Univ. Press, 2013.