Miroslav Dudík Daniel Hsu Satyen Kale Nikos Karampatziakis John Langford Lev Reyzin Tong Zhang 0001 Efficient Optimal Learning for Contextual Bandits http://arxiv.org/abs/1106.2369 2011 CoRR abs/1106.2369 db/journals/corr/corr1106.html#abs-1106-2369