Machine learning has made remarkable advancements, but confidently utilising learning-enabled components in safety-critical domains still poses challenges. Among the challenges, it is known that a rigorous, yet practical, way of achieving safety guarantees is one of the most prominent. In this paper, we first discuss the engineering and research challenges associated with the design and verification of such systems. Then, based on the observation that existing works cannot actually achieve provable guarantees, we promote a two-step verification method for the ultimate achievement of provable statistical guarantees.

翻译：机器学习已取得显著进展，但在安全关键领域自信地使用学习赋能组件仍面临挑战。其中，实现严格且实用的安全保障方法是最突出的挑战之一。本文首先探讨了此类系统设计与验证相关的工程与研究挑战。随后，基于现有工作实际上无法实现可证明保证的观察，我们提出了一种两步验证方法，以最终实现可证明的统计保证。