Explaining Control Policies through Predicate Decision Diagrams
This work addresses explainability in automated controllers for complex systems, offering a domain-specific improvement over existing methods.
The paper tackles the problem of explaining safety-critical controllers by introducing Predicate Decision Diagrams (PDDs), which combine decision trees and binary decision diagrams to reduce size and improve interpretability, resulting in a synthesis pipeline for efficient construction.
Safety-critical controllers of complex systems are hard to construct manually. Automated approaches such as controller synthesis or learning provide a tempting alternative but usually lack explainability. To this end, learning decision trees (DTs) have been prevalently used towards an interpretable model of the generated controllers. However, DTs do not exploit shared decision-making, a key concept exploited in binary decision diagrams (BDDs) to reduce their size and thus improve explainability. In this work, we introduce predicate decision diagrams (PDDs) that extend BDDs with predicates and thus unite the advantages of DTs and BDDs for controller representation. We establish a synthesis pipeline for efficient construction of PDDs from DTs representing controllers, exploiting reduction techniques for BDDs also for PDDs.