Copeland dueling bandits
A version of the dueling bandit problem is addressed in which a Condorcet winner may not exist. Two algorithms are proposed that instead seek to minimize regret with respect to the Copeland winner, which, unlike the Condorcet winner, is guaranteed to exist. The first, Copeland Confidence Bound (CCB)...
Hlavní autoři: | Zoghi, M, Karnin, Z, Whiteson, S, Rijke, M |
---|---|
Médium: | Conference item |
Vydáno: |
2015
|
Podobné jednotky
-
Melancholic Mem in the Third Life of Grange Copeland
Autor: Sedehi, Kamelia Talebian, a další
Vydáno: (2015) -
Good Outcome Following Copeland Hemiarthroplasty for Acromegalic Arthropathy
Autor: S. E. Johnson-Lynn, a další
Vydáno: (2011-01-01) -
Synergy in science: an interview with Neal Copeland and Nancy Jenkins
Vydáno: (2012-11-01) -
Exponential Regret Bounds for Gaussian Process Bandits with Deterministic Observations
Autor: de Freitas, N, a další
Vydáno: (2012) -
StreamingBandit: Experimenting with Bandit Policies
Autor: Jules Kruijswijk, a další
Vydáno: (2020-08-01)