Entropy-Guided Control Improvisation
This work addresses the need for unpredictable controllers in domains like patrolling and testing, offering a novel synthesis method that is incremental over existing declarative constraint approaches.
The authors tackled the problem of synthesizing control policies with specified randomness, introducing the ERCI framework that supports hard, soft, and randomization constraints using causal entropy. They demonstrated that this approach remains tractable theoretically and empirically, extending support for adversarial and probabilistic uncertainty.
High level declarative constraints provide a powerful (and popular) way to define and construct control policies; however, most synthesis algorithms do not support specifying the degree of randomness (unpredictability) of the resulting controller. In many contexts, e.g., patrolling, testing, behavior prediction,and planning on idealized models, predictable or biased controllers are undesirable. To address these concerns, we introduce the \emph{Entropic Reactive Control Improvisation} (ERCI) framework and algorithm which supports synthesizing control policies for stochastic games that are declaratively specified by (i) a \emph{hard constraint} specifying what must occur, (ii) a \emph{soft constraint} specifying what typically occurs, and (iii) a \emph{randomization constraint} specifying the unpredictability and variety of the controller, as quantified using causal entropy. This framework, extends the state of the art by supporting arbitrary combinations of adversarial and probabilistic uncertainty in the environment. ERCI enables a flexible modeling formalism which we argue, theoretically and empirically, remains tractable.