A new algorithm, GP-UCB-CPD, achieves sub-linear cumulative regret for non-stationary X-armed bandits with unknown change points, matching the stationary GP-UCB rate when the minimum change magnitude is fixed.
The financing of innovation : learning and stopping
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
How to gamble with non-stationary $\mathcal{X}$-armed bandits and have no regrets
A new algorithm, GP-UCB-CPD, achieves sub-linear cumulative regret for non-stationary X-armed bandits with unknown change points, matching the stationary GP-UCB rate when the minimum change magnitude is fixed.