However, in adversarial environments, these methods can be non-robust, since the return is dependent on the strategies of both the decision-maker and adversary. Training a probabilistic model conditioned on observed return to predict action can fail to generalize, as the trajectories that achieve a return in the dataset might have done so due to a suboptimal behavior adversary.

During initial iterations of training in most Reinforcement Learning (RL) algorithms, agents perform asignificant number ofrandom exploratory steps.

In the stochastic linear contextual bandit setting there exist several minimax procedures for exploration with policies that are reactive to the data being acquired.

Toour knowledge, such guarantees in this setting are providedforthefirsttime.

Toour knowledge, such guarantees in this setting are providedforthefirsttime.

In complex systems, we often observe complex global behavior emerge from a collection of agents interacting with each other in their environment, with each individual agent acting only on locally available information, without knowing thefullpicture.