Learning a Behavior Model of Hybrid Systems Through Combining Model-Based Testing and Machine Learning (Full Version)
Learning a Behavior Model of Hybrid Systems Through Combining Model-Based Testing and Machine Learning (Full Version)
复制标题
通过结合基于模型的测试和机器学习来学习混合系统的行为模型（完整版）
DOI：
--
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Markus Tranninger
中科院分区：
文献类型：
--
作者：
B. Aichernig;R. Bloem;M. Ebrahimi;M. Horn;F. Pernkopf;Wolfgang Roth;Astrid Rupp;Martin Tappler;Markus Tranninger
模型在信息物理系统的设计过程中起着至关重要的作用。它们构成了仿真和分析的基础，并有助于尽早识别设计问题。然而，构建包括物理和数字行为的模型是具有挑战性的。因此，人们对通过机器学习来学习这种混合行为非常感兴趣，机器学习需要足够的和有代表性的训练数据来充分覆盖物理系统的行为。在这项工作中，我们利用自动机学习和基于模型的测试相结合，完全自动地生成足够的训练数据。
队列场景的实验结果表明，与从随机生成的数据中学习的模型相比，使用这些数据学习的递归神经网络获得了更好的结果。特别是，碰撞检测的分类误差减少了五分之一，并且在训练样本少三个数量级的情况下获得了类似的F1分数。
Models play an essential role in the design process of cyber-physical systems. They form the basis for simulation and analysis and help in identifying design problems as early as possible. However, the construction of models that comprise physical and digital behavior is challenging. Therefore, there is considerable interest in learning such hybrid behavior by means of machine learning which requires sufficient and representative training data covering the behavior of the physical system adequately. In this work, we exploit a combination of automata learning and model-based testing to generate sufficient training data fully automatically.
Experimental results on a platooning scenario show that recurrent neural networks learned with this data achieved significantly better results compared to models learned from randomly generated data. In particular, the classification error for crash detection is reduced by a factor of five and a similar F1-score is obtained with up to three orders of magnitude fewer training samples.