Tracking the Best of Many Experts

By András György, Tamás Linder, Gábor Lugosi. 2005

31 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Decentralized Online Learning: Take Benefits from Others’ Data without Sharing Your Own to Track Global Trend (2019, background): “Shahrampour and Jadbabaie [2018] provides a dynamic regret (defined in Eq.”
Online Multi-task Learning with Hard Constraints (2009, background): “Recall th at in the online shortest path problem (see, e.g., [TW04, GLL04, GLL05 ]) the decision maker selects, at each round of the game, a path between two given vertices (the source and the sink) in a given graph.”
Proximal Online Gradient is Optimum for Dynamic Regret (2018, background): “3 Shifting regret (or tracking regret) The M−shifting regret of an algorithm A ∈ A is defined by [6, 38, 40–47] R̃T := T ∑”
Prediction, learning, and games (2006, background): “5 is based on György, Linder, and Lugosi [139].”
Mirror Descent Meets Fixed Share (and feels no regret) (2012, background): “This problem is often called “tracking the best expert” —see, e.g., [4, 5, 1, 6, 7], and it is well known that exponential weights with weight sharing, which corresponds to the ﬁxed-share algorithm of [4], achieves a good shifting bound in this setting.”
The On-Line Shortest Path Problem Under Partial Monitoring (2007, background): “Keywords: On-line learning, shortest path problem, multi-armed bandit problem”
A new look at shifting regret (2012, background): “…is often called “tracking the best expert” —see, e.g.,Herbster and Warmuth (1998); Vovk (1999); Herbster and Warmuth(2001); Bousquet and Warmuth(2002); György et al. (2005), and it is well known that EWA with weight sharing, which corresponds to the fixedshare algorithm ofHerbster and…”