Compositional falsification of cyber-physical systems with machine learning components

DOI10.1007/s10817-018-09509-5zbMath1468.68126arXiv1703.00978OpenAlexW2604347212WikidataQ128537381 ScholiaQ128537381MaRDI QIDQ2331078

Alexandre Donzé, Tommaso Dreossi, Sanjit A. Seshia

Publication date: 25 October 2019

Published in: Journal of Automated Reasoning (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1703.00978

zbMATH Keywords

neural networkstemporal logicmachine learningcyber-physical systemsdeep learningfalsificationautonomous driving

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Learning and adaptive systems in artificial intelligence (68T05) Control/observation systems involving computers (process control, etc.) (93C83) Specification and verification (program logics, model checking, etc.) (68Q60)

Related Items

Adaptive testing for specification coverage and refinement in CPS models ⋮ Compositional falsification of cyber-physical systems with machine learning components

Uses Software

• S-TaLiRo
• TensorFlow
• Caffe
• C2e2
• Breach
• DeepFool
• ImageNet
• GitHub
• SqueezeNet
• self-driving-car-sim
• AlexNet
• DeepXplore

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Low-discrepancy and low-dispersion sequences
• Selection of relevant features and examples in machine learning
• DeepXplore
• Analysis of classifiers' robustness to adversarial perturbations
• Safety verification of deep neural networks
• Logical clustering and learning for time-series data
• Compositional falsification of cyber-physical systems with machine learning components
• S-TaLiRo: A Tool for Temporal Logic Falsification for Hybrid Systems
• Quasi-Random Sequences and Their Discrepancies
• Formal Techniques, Modelling and Analysis of Timed and Fault-Tolerant Systems
• Geometric discrepancy. An illustrated guide
• Formal specification for deep neural networks