Logical Clustering and Learning for Time-Series Data

Vazquez-Chanlatte, Marcell; Deshmukh, Jyotirmoy V.; Jin, Xiaoqing; Seshia, Sanjit A.

doi:10.1007/978-3-319-63387-9_15

Cited by 38 publications

(58 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While this is useful, it is arguably even more useful to obtain higher-level interpretable insight into where the training data falls short, what new scenarios must be added to the training set, and how the learning algorithms' parameters must be adjusted to improve accuracy. For example, one could use techniques for mining specifications or requirements (e.g., [21,42]) to aggregate interesting test images or video into a cluster that can be represented in a high-level fashion. One could also apply our ML Analyzer outside the falsification context, such as for controller synthesis.…”

Section: Resultsmentioning

confidence: 99%

Compositional Falsification of Cyber-Physical Systems with Machine Learning Components

Donzé²,

2019

Self Cite

View full text Add to dashboard Cite

Cyber-physical systems (CPS), such as automotive systems, are starting to include sophisticated machine learning (ML) components. Their correctness, therefore, depends on properties of the inner ML modules. While learning algorithms aim to generalize from examples, they are only as good as the examples provided, and recent efforts have shown that they can produce inconsistent output under small adversarial perturbations. This raises the question: can the output from learning components lead to a failure of the entire CPS? In this work, we address this question by formulating it as a problem of falsifying signal temporal logic specifications for CPS with ML components. We propose a compositional falsification framework where a temporal logic falsifier and a machine learning analyzer cooperate with the aim of finding falsifying executions of the considered model. The efficacy of the proposed technique is shown on an automatic emergency braking system model with a perception component based on deep neural networks.

show abstract

Section: Resultsmentioning

confidence: 99%

Compositional Falsification of Cyber-Physical Systems with Machine Learning Components

Donzé²,

2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…Once mapped into this space, the signals can be subject to various learning and clustering algorithms as suggested recently in [41].…”

Section: Discussionmentioning

confidence: 99%

Efficient Parametric Identification for STL

Bakhirkin

Ferrère

Maler

2018

Proceedings of the 21st International Conference on Hybrid Systems: Computation and Control (Part of CPS Week)

View full text Add to dashboard Cite

We describe a new algorithm for the parametric identification problem for signal temporal logic (STL), stated as follows. Given a densetime real-valued signal w and a parameterized temporal logic formula φ, compute the subset of the parameter space that renders the formula satisfied by the signal. Unlike previous solutions, which were based on search in the parameter space or quantifier elimination, our procedure works recursively on φ and computes the evolution over time of the set of valid parameter assignments. This procedure is similar to that of monitoring or computing the robustness of φ relative to w. Our implementation and experiments demonstrate that this approach can work well in practice.

show abstract

“…Finally, preliminary exploration of this idea appeared in prior work by some of the co-authors in [16]. The key difference is the previous work required users to provide a ranking of parameters appearing in a signal predicate, in order to project time-series data to unique points in the parameter space.…”

Section: Related Workmentioning

confidence: 99%

“…While powerful, translating the knowledge of domain-specific experts into features remains a non-trivial endeavor. More recently, it has been shown that Parametric Signal Temporal Logic formula along with a total ordering on the parameter space can be used to extract feature vectors for learning temporal logical predicates characterizing driving patterns, overshoot of diesel engine re-flow rates, and grading for simulated robot controllers in a Massively Open Online Course [16]. Crucially, the technique of learning through the lens of a logical formula means that learned artifacts can be readily leveraged by existing formal methods infrastructure for verification, synthesis, falsification, and monitoring.…”

Section: Introductionmentioning

confidence: 99%

Time-Series Learning Using Monotonic Logical Properties

Vazquez-Chanlatte

Ghosh

Deshmukh

et al. 2018

Runtime Verification

Self Cite

View full text Add to dashboard Cite

Cyber-physical systems of today are generating large volumes of time-series data. As manual inspection of such data is not tractable, the need for learning methods to help discover logical structure in the data has increased. We propose a logic-based framework that allows domain-specific knowledge to be embedded into formulas in a parametric logical specification over time-series data. The key idea is to then map a time series to a surface in the parameter space of the formula. Given this mapping, we identify the Hausdorff distance between boundaries as a natural distance metric between two time-series data under the lens of the parametric specification. This enables embedding non-trivial domain-specific knowledge into the distance metric and then using offthe-shelf machine learning tools to label the data. After labeling the data, we demonstrate how to extract a logical specification for each label. Finally, we showcase our technique on real world traffic data to learn classifiers/monitors for slow-downs and traffic jams.

show abstract

Logical Clustering and Learning for Time-Series Data

Cited by 38 publications

References 22 publications

Compositional Falsification of Cyber-Physical Systems with Machine Learning Components

Compositional Falsification of Cyber-Physical Systems with Machine Learning Components

Efficient Parametric Identification for STL

Time-Series Learning Using Monotonic Logical Properties

Contact Info

Product

Resources

About