TY - GEN
T1 - Calibration and internal no-regret with random signals
AU - Perchet, Vianney
PY - 2009/12/1
Y1 - 2009/12/1
N2 - A calibrated strategy can be obtained by performing a strategy that has no internal regret in some auxiliary game. Such a strategy can be constructed explicitly with the use of Blackwell's approachability theorem, in an other auxiliary game. We establish the converse: a strategy that approaches a convex B-set can be derived from the construction of a calibrated strategy. We develop these tools in the framework of a game with partial monitoring, where players do not observe the actions of their opponents but receive random signals, to define a notion of internal regret and construct strategies that have no such regret.
AB - A calibrated strategy can be obtained by performing a strategy that has no internal regret in some auxiliary game. Such a strategy can be constructed explicitly with the use of Blackwell's approachability theorem, in an other auxiliary game. We establish the converse: a strategy that approaches a convex B-set can be derived from the construction of a calibrated strategy. We develop these tools in the framework of a game with partial monitoring, where players do not observe the actions of their opponents but receive random signals, to define a notion of internal regret and construct strategies that have no such regret.
U2 - 10.1007/978-3-642-04414-4_10
DO - 10.1007/978-3-642-04414-4_10
M3 - Conference contribution
AN - SCOPUS:77952069205
SN - 3642044131
SN - 9783642044137
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 68
EP - 82
BT - Algorithmic Learning Theory - 20th International Conference, ALT 2009, Proceedings
T2 - 20th International Conference on Algorithmic Learning Theory, ALT 2009
Y2 - 3 October 2009 through 5 October 2009
ER -