“…A number of methods have been proposed for online reachability analysis that rely on separating the reachability computation into distinct offline and online phases. However, these methods are limited to restricted classes of models [10], or require handcrafted optimization of the HA's derivatives [4], or are efficient only for low-dimensional systems and simple dynamics [25]. In contrast, NSC [21] is based on learning DNN-based classifiers, is fully automated and has negligible computational cost at runtime.…”
Section: Related Workmentioning
confidence: 99%
“…Our focus is on the online analysis of hybrid systems and, in particular, on the predictive monitoring (PM) problem [10]; i.e., the problem of predicting, at runtime, whether or not an unsafe state can be reached from the current system state within a given time bound. PM is at the core of architectures for runtime safety assurance such as Simplex [26], where the system switches to a safe fallback mode whenever PM indicates the potential for an imminent failure.…”
Neural State Classification (NSC) is a recently proposed method for runtime predictive monitoring of Hybrid Automata (HA) using deep neural networks (DNNs). NSC trains a DNN as an approximate reachability predictor that labels a given HA state x as positive if an unsafe state is reachable from x within a given time bound, and labels x as negative otherwise. NSC predictors have very high accuracy, yet are prone to prediction errors that can negatively impact reliability. To overcome this limitation, we present Neural Predictive Monitoring (NPM), a technique based on NSC and conformal prediction that complements NSC predictions with statistically sound estimates of uncertainty. This yields principled criteria for the rejection of predictions likely to be incorrect, without knowing the true reachability values. We also present an active learning method that significantly reduces both the NSC predictor's error rate and the percentage of rejected predictions. Our approach is highly efficient, with computation times on the order of milliseconds, and effective, managing in our experimental evaluation to successfully reject almost all incorrect predictions.
“…A number of methods have been proposed for online reachability analysis that rely on separating the reachability computation into distinct offline and online phases. However, these methods are limited to restricted classes of models [10], or require handcrafted optimization of the HA's derivatives [4], or are efficient only for low-dimensional systems and simple dynamics [25]. In contrast, NSC [21] is based on learning DNN-based classifiers, is fully automated and has negligible computational cost at runtime.…”
Section: Related Workmentioning
confidence: 99%
“…Our focus is on the online analysis of hybrid systems and, in particular, on the predictive monitoring (PM) problem [10]; i.e., the problem of predicting, at runtime, whether or not an unsafe state can be reached from the current system state within a given time bound. PM is at the core of architectures for runtime safety assurance such as Simplex [26], where the system switches to a safe fallback mode whenever PM indicates the potential for an imminent failure.…”
Neural State Classification (NSC) is a recently proposed method for runtime predictive monitoring of Hybrid Automata (HA) using deep neural networks (DNNs). NSC trains a DNN as an approximate reachability predictor that labels a given HA state x as positive if an unsafe state is reachable from x within a given time bound, and labels x as negative otherwise. NSC predictors have very high accuracy, yet are prone to prediction errors that can negatively impact reliability. To overcome this limitation, we present Neural Predictive Monitoring (NPM), a technique based on NSC and conformal prediction that complements NSC predictions with statistically sound estimates of uncertainty. This yields principled criteria for the rejection of predictions likely to be incorrect, without knowing the true reachability values. We also present an active learning method that significantly reduces both the NSC predictor's error rate and the percentage of rejected predictions. Our approach is highly efficient, with computation times on the order of milliseconds, and effective, managing in our experimental evaluation to successfully reject almost all incorrect predictions.
“…Even though research on reachability checking of hybrid systems [13,1] has produced effective verification algorithms and tools [10,7,11], comparably little has been done to make these algorithms efficient for online analysis. Existing approaches are limited to restricted classes of models [8], or require handcrafted optimization of the HA's derivatives [2], or are efficient only for low-dimensional systems and simple dynamics [21]. NSC [19] (introduced in Section 2) overcomes these limitations because, by employing machine learning models, it is fully automated and its performance is not affected by the model size or complexity.…”
Neural State Classification (NSC) [19] is a scalable method for the analysis of hybrid systems, which consists in learning a neural network-based classifier able to detect whether or not an unsafe state can be reached from a certain configuration of a hybrid system. NSC has very high accuracy, yet it is prone to prediction errors that can affect system safety. To overcome this limitation, we present a method, based on the theory of conformal prediction, that complements NSC predictions with statistically sound estimates of prediction uncertainty. This results in a principled criterion to reject potentially erroneous predictions a priori, i.e., without knowing the true reachability values. Our approach is highly efficient (with runtimes in the order of milliseconds) and effective, managing in our experiments to successfully reject almost all the wrong NSC predictions.
“…Note CORA Since the dynamics of this example is dominated by the input after one second, we use the step size 0.002 for t ∈ [0, 1] and the step size 0.01 for t ∈ [1,20]. The zonotope order is chosen as 100. .…”
Section: Resultsmentioning
confidence: 99%
“…Unlike convex set representations, symbolic flowpipes are usually more time-costly to obtain, however, they are only ODE related and can be directly reused in a safety verification task, i.e., with a different initial set or unsafe condition. Besides, symbolic flowpipes can be used in generating relational abstractions [30,19] and real-time monitoring [20] for dynamical systems. In the current version, Flow* simply treats all real numbers as intervals in order to take roundoff errors into account in all computational tasks.…”
This report presents the results of a friendly competition for formal verification of continuous and hybrid systems with linear continuous dynamics. The friendly competition took place as part of the workshop Applied Verification for Continuous and Hybrid Systems (ARCH) in 2018. In its second edition, 9 tools have been applied to solve six different benchmark problems in the category for linear continuous dynamics (in alphabetical order): CORA, CORA/SX, C2E2, Flow*, HyDRA, Hylaa, Hylaa-Continuous, JuliaReach, SpaceEx, and XSpeed. This report is a snapshot of the current landscape of tools and the types of benchmarks they are particularly suited for. Due to the diversity of problems, we are not ranking tools, yet the presented results probably provide the most complete assessment of tools for the safety verification of continuous and hybrid systems with linear continuous dynamics up to this date.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.