Information-Gathering in Latent Bandits

Alexander Galozy, Slawomir Nowaczyk

In the latent bandit problem, the learner has access to reward distributions and -- for the non-stationary variant -- transition models of the environment. The reward distributions…