Survey on mining signal temporal logic specifications

Bartocci, Ezio; Mateis, Cristinel; Nesterini, Eleonora; Ničković, Dejan

doi:10.1016/j.ic.2022.104957

Cited by 15 publications

(5 citation statements)

References 71 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Logic inference Logic inference (Bartocci et al 2022) is the process of generating logic properties based on observed facts when the desired system property is unknown or only partially available. Given some prior knowledge about the possible form of a logic property, the logic inference algorithm (specification mining (Jha et al 2017)) learns the complete logic formula.…”

Section: Logic Inference Through Observed Datamentioning

confidence: 99%

Formal Logic Enabled Personalized Federated Learning through Property Inference

An,

Johnson,

2024

AAAI

View full text Add to dashboard Cite

Recent advancements in federated learning (FL) have greatly facilitated the development of decentralized collaborative applications, particularly in the domain of Artificial Intelligence of Things (AIoT). However, a critical aspect missing from the current research landscape is the ability to enable data-driven client models with symbolic reasoning capabilities. Specifically, the inherent heterogeneity of participating client devices poses a significant challenge, as each client exhibits unique logic reasoning properties. Failing to consider these device-specific specifications can result in critical properties being missed in the client predictions, leading to suboptimal performance. In this work, we propose a new training paradigm that leverages temporal logic reasoning to address this issue. Our approach involves enhancing the training process by incorporating mechanically generated logic expressions for each FL client. Additionally, we introduce the concept of aggregation clusters and develop a partitioning algorithm to effectively group clients based on the alignment of their temporal reasoning properties. We evaluate the proposed method on two tasks: a real-world traffic volume prediction task consisting of sensory data from fifteen states and a smart city multi-task prediction utilizing synthetic data. The evaluation results exhibit clear improvements, with performance accuracy improved by up to 54% across all sequential prediction models.

show abstract

Section: Logic Inference Through Observed Datamentioning

confidence: 99%

Formal Logic Enabled Personalized Federated Learning through Property Inference

An,

Johnson,

2024

AAAI

View full text Add to dashboard Cite

show abstract

“…Alessio Cecconi et al [33] proposed a comprehensive measurement framework to solve the problems of a lack of feedback and scalability in declarative process mining. Ezio Bartocci et al [34] summarize the existing methods for extracting specifications from cyber-physical systems (CPS) based on supervised versus unsupervised learning.…”

Section: Mining Specifications Expressed In Temporal Logic Formulasmentioning

confidence: 99%

Specification Mining Based on the Ordering Points to Identify the Clustering Structure Clustering Algorithm and Model Checking

Fan,

Wang

2024

Algorithms

View full text Add to dashboard Cite

Software specifications are of great importance to improve the quality of software. To automatically mine specifications from software systems, some specification mining approaches based on finite-state automatons have been proposed. However, these approaches are inaccurate when dealing with large-scale systems. In order to improve the accuracy of mined specifications, we propose a specification mining approach based on the ordering points to identify the clustering structure clustering algorithm and model checking. In the approach, the neural network model is first used to produce the feature values of states in the traces of the program. Then, according to the feature values, finite-state automatons are generated based on the ordering points to identify the clustering structure clustering algorithm. Further, the finite-state automaton with the highest F-measure is selected. To improve the quality of the finite-state automatons, we refine it based on model checking. The proposed approach was implemented in a tool named MCLSM and experiments, including 13 target classes, were conducted to evaluate its effectiveness. The experimental results show that the average F-measure of finite-state automatons generated by our method reaches 92.19%, which is higher than most related tools.

show abstract

“…We formalise traffic rule violation for crossing ships as matching of 𝜙 11 (Expr (12)) where the constant 𝑑𝑡 = 1 is the time step, parameter 𝜃 𝑟 ∈ [6,20] is the reaction time and 𝜃 𝑚 ∈ [6,20] is the maneuver time. The expression captures two kinds of rule violations.…”

Section: Parametric Identification For Labelled Ecg-205 ( Mini)mentioning

confidence: 99%

“…Both these approaches can be modified and applied also to PSRE but we leave this to future work. A comprehesive survey on methods for parametric identification for PSTL and general approaches that learn the structure as well as parameter values of STL formulae can be found in [12]. Now, we compare with PTA.…”

Section: Related Workmentioning

confidence: 99%

Pattern Matching and Parameter Identification for Parametric Timed Regular Expressions

Mambakam

Asarin

Basset

et al. 2023

Proceedings of the 26th ACM International Conference on Hybrid Systems: Computation and Control

View full text Add to dashboard Cite

Timed formalisms such as Timed Automata (TA), Signal Temporal Logic (STL) and Timed Regular expressions (TRE) have been previously applied as behaviour specifications for monitoring or runtime verification, in particular, under the form of pattern-matching, i.e. computing the set of all the segments of a given system run that satisfy the specification.In this work, timed regular expressions with parameters (for timing delays and for signal values) are considered. We define several classes of parametric expressions (based on Boolean or real-valued signals and discrete events), and tackle the problem of computing a parametric match-set, i.e. the parameter values and time segments of data that give a match for a given expression. We propose efficient data structures for representing match-sets (combining zones and polytopes), and devise pattern-matching algorithms. All these different types and algorithms are combined into a single implementation under a tool named parameTRE. We illustrate the approach on several examples, from electrocardiograms to driving patterns. CCS CONCEPTS• Theory of computation → Timed and hybrid models; • Computing methodologies → Algebraic algorithms.

show abstract

Survey on mining signal temporal logic specifications

Cited by 15 publications

References 71 publications

Formal Logic Enabled Personalized Federated Learning through Property Inference

Formal Logic Enabled Personalized Federated Learning through Property Inference

Specification Mining Based on the Ordering Points to Identify the Clustering Structure Clustering Algorithm and Model Checking

Pattern Matching and Parameter Identification for Parametric Timed Regular Expressions

Contact Info

Product

Resources

About