Sciweavers

Share
ALT
2009
Springer

The Follow Perturbed Leader Algorithm Protected from Unbounded One-Step Losses

8 years 11 months ago
The Follow Perturbed Leader Algorithm Protected from Unbounded One-Step Losses
In this paper the sequential prediction problem with expert advice is considered for the case when the losses of experts suffered at each step can be unbounded. We present some modification of Kalai and Vempala algorithm of following the perturbed leader where weights depend on past losses of the experts. New notions of a volume and a scaled fluctuation of a game are introduced. We present an algorithm protected from unrestrictedly large one-step losses. This algorithm has the optimal performance in the case when the scaled fluctuations of onestep losses of experts of the pool tend to zero.
Vladimir V. V'yugin
Added 23 Jul 2010
Updated 23 Jul 2010
Type Conference
Year 2009
Where ALT
Authors Vladimir V. V'yugin
Comments (0)
books