Trust in a decision-making system requires both safety guarantees and the ability to interpret and understand its behavior. This is particularly important for learned systems, whose decision-making processes are often highly opaque. Shielding is a prominent model-based technique for enforcing safety in reinforcement learning. However, because shields are automatically synthesized using rigorous formal methods, their decisions are often similarly difficult for humans to interpret. Recently, decision trees became customary to represent controllers and policies. However, since shields are inherently non-deterministic, their decision tree representations become too large to be explainable in practice. To address this challenge, we propose a novel approach for explainable safe RL that enhances trust by providing human-interpretable explanations of the shield's decisions. Our method represents the shielding policy as a hierarchy of decision trees, offering top-down, case-based explanations. At design time, we use a world model to analyze the safety risks of executing actions in given states. Based on this risk analysis, we construct both the shield and a high-level decision tree that classifies states into risk categories (safe, critical, dangerous, unsafe), providing an initial explanation of why a given situation may be safety-critical. At runtime, we generate localized decision trees that explain which actions are allowed and why others are deemed unsafe. Altogether, our method facilitates the explainability of the safety aspect in the safe-by-shielding reinforcement learning. Our framework requires no additional information beyond what is already used for shielding, incurs minimal overhead, and can be readily integrated into existing shielded RL pipelines. In our experiments, we compute explanations using decision trees that are several orders of magnitude smaller than the original shield.

Abstract The behavior of neural networks (NNs) on previously unseen types of data (out-of-distribution or OOD) is typically unpredictable. This can be dangerous if the network’s output is used for decision making in a safety-critical system. Hence, detecting that an input is OOD is crucial for the safe application of the NN. Verification approaches do not scale to practical NNs, making runtime monitoring more appealing for practical use. While various monitors have been suggested recently, their optimization for a given problem, as well as comparison with each other and reproduction of results, remain challenging. We present a tool for users and developers of NN monitors. It allows for (i) application of various types of monitors from the literature to a given input NN, (ii) optimization of the monitor’s hyperparameters, and (iii) experimental evaluation and comparison to other approaches. Besides, it facilitates the development of new monitoring approaches. We demonstrate the tool’s usability on several use cases of different types of users as well as on a case study comparing different approaches from recent literature.

Abstract We provide a learning-based technique for guessing a winning strategy in a parity game originating from an LTL synthesis problem. A cheaply obtained guess can be useful in several applications. Not only can the guessed strategy be applied as best-effort in cases where the game’s huge size prohibits rigorous approaches, but it can also increase the scalability of rigorous LTL synthesis in several ways. Firstly, checking whether a guessed strategy is winning is easier than constructing one. Secondly, even if the guess is wrong in some places, it can be fixed by strategy iteration faster than constructing one from scratch. Thirdly, the guess can be used in on-the-fly approaches to prioritize exploration in the most fruitful directions. In contrast to previous works, we (i) reflect the highly structured logical information in game’s states, the so-called semantic labelling, coming from the recent LTL-to-automata translations, and (ii) learn to reflect it properly by learning from previously solved games, bringing the solving process closer to human-like reasoning.