arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2605.07541eess.SYcs.SY
收藏

学习基于神经的混合代理以进行基于梯度的验证

Learning Neural Hybrid Surrogates for Gradient-Based Falsification

Chalmers University of Technology(查尔姆斯理工大学)

Lasse Kötz, Knut Åkesson

AI总结:

本文提出基于代理的验证方法,通过学习可微混合自动机模型来验证混合动态系统,提升了对离散转换的处理能力,并在多个基准测试中表现出更高的样本效率。

摘要:

混合动态系统的验证仍然具有挑战性,因为其动态依赖于模式和离散转换。本文提出了一种基于代理的验证方法,通过从数据中学习可微混合自动机模型来实现混合系统。该方法扩展了先前的基于代理的验证方法,这些方法仅限于纯连续动态。具体而言,我们使用神经混合自动机来学习潜在模式编码器和对应的模式条件向量场。一旦代理将每个模式与一个关联的向量场配对,就会使用现有轨迹数据推断转换守卫。学习到的代理随后会受到基于梯度的最优控制公式的影响,以最小化安全规范的平滑近似值来发现安全违规。最后,通过在原始系统上执行最优控制解决方案的实验来确保正确性。所提出的方法在大多数评估的基准规范上一致地发现了反例;在这些情况下,它在使用减少的模拟预算时实现了竞争或改进的样本效率。

英文摘要:

Falsification of hybrid dynamical systems remains challenging due to mode-dependent dynamics and discrete transitions. In this work, we propose a surrogate-based falsification approach that enables hybrid systems by learning a differentiable hybrid automaton model from data. This extends previous surrogate-based falsification methods, which were limited to purely continuous dynamics. Specifically, we employ neural hybrid automata to learn both a latent mode encoder and the corresponding mode-conditioned vector fields. Once the surrogate has paired each mode with an associated vector field, the transition guards are inferred using existing trajectory data. The learned surrogate is subsequently subjected to a gradient-based optimal control formulation, which minimizes a smooth approximation of the safety specification to find safety violations. In the last step, an experiment with the optimal control solution is carried out on the original system to ensure soundness. The proposed method consistently uncovers counterexamples on a majority of evaluated benchmark specifications; on these cases, it achieves competitive or improved sample efficiency than other tools while using a reduced simulation budget.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑