# Kernel-based methods for bandit convex optimization

11 Jul 2016Sébastien BubeckRonen EldanYin Tat Lee

We consider the adversarial convex bandit problem and we build the first $\mathrm{poly}(T)$-time algorithm with $\mathrm{poly}(n) \sqrt{T}$-regret for this problem. To do so we introduce three new ideas in the derivative-free optimization literature: (i) kernel methods, (ii) a generalization of Bernoulli convolutions, and (iii) a new annealing schedule for exponential weights (with increasing learning rate)... (read more)

PDF Abstract

No code implementations yet. Submit your code now