%ONE
%We investigate the use of Reward Machines for Deep RL in noisy and uncertain enviroments.

TWO
We investigate the use of Reward Machines in deep RL under an uncertain interpretation of the domain-specific vocabulary.

THREE
We investigate the use of Reward Machines in deep RL under an uncertain interpretation of the vocabulary


----------------

OLD
We investigate whether RL agents can follow formal language reward specifications under an uncertain symbolic interpretation.