# Efficient Online Bandit Multiclass Learning with O(sqrt{T}) Regret

Alina BeygelzimerFrancesco OrabonaChicheng Zhang

We present an efficient second-order algorithm with $\tilde{O}(1/\eta \sqrt{T})$ regret for the bandit online multiclass problem. The regret bound holds simultaneously with respect to a family of loss functions parameterized by $\eta$, ranging from hinge loss ($\eta=0$) to squared hinge loss ($\eta=1$)... (read more)

PDF Abstract