Partial monitoring - Classification, regret bounds, and algorithms

Mathematics of Operations Research

Volumn 39, Issue 4, 2014, Pages 967-997

  Bartók, Gábor ^a   Foster, Dean P ^b   Pál, Dávid ^c   Rakhlin, Alexander ^d   Szepesvári, Csaba ^e  

^a ETH ZURICH   (Switzerland)

^b YAHOO RESEARCH   (United States)

^c GOOGLE INC   (United States)

^d UNIVERSITY OF PENNSYLVANIA   (United States)

^e UNIVERSITY OF ALBERTA   (Canada)

Author keywords

Imperfect information; Partial monitoring; Regret analysis; Repeated games

Indexed keywords

COMPUTER GAMES;

COMPUTATIONALLY EFFICIENT; FEEDBACK SIGNAL; IMPERFECT INFORMATION; MINIMAX REGRET; REGRET ANALYSIS; REGRET BOUNDS; REPEATED GAMES; UPPER BOUND;

LEARNING ALGORITHMS;

EID: 84908695406     PISSN: 0364765X     EISSN: 15265471     Source Type: Journal    
DOI: 10.1287/moor.2014.0663     Document Type: Article

References (27)

1. Abernethy J, Rakhlin A (2009) Beating the adaptive bandit with high probability. Proc. 22nd Conf. Learn. Theory, COLT '09 (Omnipress, Madison, WI).
2. Abernethy J, Hazan E, Rakhlin A (2008) Competing in the dark: An efficient algorithm for bandit linear optimization. Proc. 21st Conf. Learn. Theory COLT '08 (Omnipress, Madison, WI), 263-273.
3. Agarwal A, Bartlett P, Dama M (2010) Optimal allocation strategies for the dark pool problem. Proc. 13th Internat. Conf. Artificial Intelligence Statist. AISTATS '10, Sardinia, Italy, 9-16.
4. Antos A, Bartók G, Pál D, Szepesvári Cs (2013) Toward a classification of finite partial-monitoring games. Theory Comput. Sci. 473:77-99.
5. Audibert J-Y, Bubeck S (2009) Minimax policies for adversarial and stochastic bandits. Proc. 22nd Conf. Learn. Theory, COLT '09 (Omnipress, Madison, WI).
6. Auer P, Cesa-Bianchi N, Freund Y, Schapire RE (2002) The nonstochastic multiarmed bandit problem. SIAM J. Comput. 32(1):48-77.
7. Bartók G (2013) A near-optimal algorithm for finite partial-monitoring games against adversarial opponents. Proc. 26th Conf. Learn. Theory, COLT '13, Princeton, New Jersey, 696-710.
8. Bartók G, Szepesvári Cs (2012) Partial monitoring with side information. Bshouty NH, Stoltz G, Vayatis N, Zeugmann T, eds. Proc. 23rd Internat. Conf. Algorithmic Learn. Theory, ALT '12 (Springer, Berlin, Heidelberg), 305-319.
9. Bartók G, Pál D, Szepesvári C (2010) Toward a classification of finite partial-monitoring games. Hutter M, Stephan F, Vovk V, Zeugmann T, eds. Proc. 21st Internat. Conf. Algorithmic Learn. Theory, ALT '10 (Springer, Berlin, Heidelberg), 224-238.
10. Bartók G, Pál D, Szepesvári C (2011) Minimax regret of finite partial-monitoring games in stochastic environments. Kakade SM, Von Luxburg U, eds. Proc. 24th Conf. Learn. Theory, COLT '11, Budapest, Hungary, 133-154.
11. Bartók G, Zolghadr N, Szepesvári Cs (2012) An adaptive algorithm for finite stochastic partial monitoring. Proc. 29th Internat. Conf. Machine Learn., ICML '12 (Omnipress, Madison, WI).
12. Blum A, Mansour Y (2007) From external to internal regret. J. Machine Learn. Res. 8:1307-1324.
13. Cesa-Bianchi N, Lugosi G, Stoltz G (2005) Minimizing regret with label efficient prediction. IEEE Trans. Inform. Theory 51(6):2152-2162.
14. Cesa-Bianchi N, Lugosi G, Stoltz G (2006) Regret minimization under partial monitoring. Math. Oper. Res. 31(3):562-580.
15. Cover TM, Thomas JA (2006) Elements of Information Theory, 2nd ed. (John Wiley & Sons, Hoboken, NJ).
16. Flaxman AD, Kalai AT, McMahan HB (2005) Online convex optimization in the bandit setting: Gradient descent without a gradient. Proc. 16th ACM-SIAM Sympos. Discrete Algorithms, SODA '05 (SIAM, Philadelphia), 385-394.
17. Foster DP, Rakhlin A (2012) No internal regret via neighborhood watch. J. Machine Learn. Res. - Proc. Track (AISTATS) 22:382-390.
18. Foster DP, Vohra RV (1997) Calibrated learning and correlated equilibrium. Games Econom. Behav. 21(1-2):40-55.
19. Kleinberg R, Leighton T (2003) The value of knowing a demand curve: Bounds on regret for online posted-price auctions. Proc. 44th IEEE Sympos. Foundations Comput. Sci. 2003, FOCS '03 (IEEE Compurter Society, Washington, DC), 594-605.
20. Littlestone N, Warmuth MK (1994) The weighted majority algorithm. Inform. Comput. 108:212-261.
21. Lugosi G, Cesa-Bianchi N (2006) Prediction, Learning, and Games (Cambridge University Press, Cambridge, UK).
22. Lugosi G, Mannor S, Stoltz G (2008) Strategies for prediction under imperfect monitoring. Math. Oper. Res 33(3):513-528.
23. Perchet V (2011) Internal regret with partial monitoring: Calibration-based optimal algorithms. J. Machine Learn. Res. 12:1893-1921.
24. Piccolboni A, Schindelhauer C (2001) Discrete prediction games with arbitrary feedback and loss. Computational Learning Theory Lecture Notes in Computer Science, Vol. 2111 (Springer, Berlin, Heidelberg), 208-223.
25. Rustichini A (1999) Minimizing regret: The general case. Games Econom. Behav. 29(1-2):224-243.
26. Vovk VG (1990) Aggregating strategies. Proc. Third Workshop on Comput. Learn. Theory, COLT '90 (Morgan Kaufmann, San Francisco), 371-383.
27. Zinkevich M (2003) Online convex programming and generalized infinitesimal gradient ascent. Proc. 20th Internat. Conf. Machine Learn., ICML '03 (AAAI Press, Palto Alto, CA), 928-936.