We present Gaptron, a randomized first-order algorithm for online multiclass classification. In the full information setting we show expected mistake bounds with respect to the logistic loss, hinge loss, and the smooth hinge loss with constant regret, where the expectation is with respect to the learner's randomness. In the bandit classification setting we show that Gaptron is the first linear time algorithm with $O(K\sqrt{T})$ expected regret, where $K$ is the number of classes. Additionally, the expected mistake bound of Gaptron does not depend on the dimension of the feature vector, contrary to previous algorithms with $O(K\sqrt{T})$ regret in the bandit classification setting. We present a new proof technique that exploits the gap between the zero-one loss and surrogate losses rather than exploiting properties such as exp-concavity or mixability, which are traditionally used to prove logarithmic or constant regret bounds.
翻译:我们展示了Gaptron(Gaptron),这是用于在线多级分类的随机一阶算法。在完整的信息设置中,我们展示了后勤损失的预期误差界限,断裂了损失,而顺滑的断断断断断断断断断断,不断后悔,对学习者随机性的期望。在土匪分类设置中,我们展示了Gaptron(Gaptron)是第一个使用$O(K\sqrt{T})的线性时间算法,预期的误差是分级数。此外,Gaptron(Gaptron)的预期误差并不取决于特性矢量的尺寸,这与以前用$O(K\sqrt{T})的算法相反。我们展示了一种新的证据技术,它利用零一损失与超额损失和超额损失之间的鸿沟,而不是利用诸如Exconity或混合性等特性,这些特性传统上用来证明对数或常数的后悔界限。