Bounded Memory Adversarial Bandits with Composite Anonymous Delayed Feedback. (arXiv:2204.12764v1 [cs.LG])

We study the adversarial bandit problem with composite anonymous delayed
feedback. In this setting, losses of an action are split into $d$ components,
spreading over consecutive rounds after the action is chosen. And in each
round, the algorithm observes the aggregation of losses that come from the
latest $d$ rounds. Previous works focus on oblivious adversarial setting, while
we investigate the harder non-oblivious setting. We show non-oblivious setting
incurs $\Omega(T)$ pseudo regret even when the loss sequence is bounded memory. …

anonymous arxiv feedback memory