Learning Interpretable Models in the Property Specification Language

We address the problem of learning human-interpretable descriptions of a complex system from a finite set of positive and negative examples of its behavior. In contrast to most of the recent work in this area, which focuses on descriptions expressed in Linear Temporal Logic (LTL), we develop a learning algorithm for formulas in the IEEE standard temporal logic PSL (Property Specification Language). Our work is motivated by the fact that many natural properties, such as an event happening at every n-th point in time, cannot be expressed in LTL, whereas it is easy to express such properties in PSL. Moreover, formulas in PSL can be more succinct and easier to interpret (due to the use of regular expressions in PSL formulas) than formulas in LTL. Our learning algorithm builds on top of an existing algorithm for learning LTL formulas. Roughly speaking, our algorithm reduces the learning task to a constraint satisfaction problem in propositional logic and then uses a SAT solver to search for a solution in an incremental fashion. We have implemented our algorithm and performed a comparative study between the proposed method and the existing LTL learning algorithm. Our results illustrate the effectiveness of the proposed approach to provide succinct human-interpretable descriptions from examples.

[1]  Alberto Camacho,et al.  Learning Interpretable Models Expressed in Linear Temporal Logic , 2019, ICAPS.

[2]  Christian Muise,et al.  Bayesian Inference of Linear Temporal Logic Specifications for Contrastive Explanations , 2019, IJCAI.

[3]  Avner Landver,et al.  The ForSpec Temporal Logic: A New Temporal Property-Specification Language , 2002, TACAS.

[4]  Eric Medvet,et al.  Automatic Synthesis of Regular Expressions from Examples , 2014, Computer.

[5]  Daniel Neider,et al.  Learning Linear Temporal Properties , 2018, 2018 Formal Methods in Computer Aided Design (FMCAD).

[6]  Fred Kröger,et al.  Temporal Logic of Programs , 1987, EATCS Monographs on Theoretical Computer Science.

[7]  Pierre Wolper Temporal Logic Can Be More Expressive , 1983, Inf. Control..

[8]  Nikolaj Bjørner,et al.  Z3: An Efficient SMT Solver , 2008, TACAS.

[9]  Rajeev Alur,et al.  Search-based program synthesis , 2018, Commun. ACM.

[10]  Heinz Riener Exact Synthesis of LTL Properties from Traces , 2019, 2019 Forum for Specification and Design Languages (FDL).

[11]  Marijn J. H. Heule,et al.  Exact DFA Identification Using SAT Solvers , 2010, ICGI.

[12]  Ivan Beschastnikh,et al.  General LTL Specification Mining (T) , 2015, 2015 30th IEEE/ACM International Conference on Automated Software Engineering (ASE).

[13]  Rajeev Alur,et al.  Syntax-guided synthesis , 2013, 2013 Formal Methods in Computer-Aided Design.

[14]  Dana Fisman,et al.  A Practical Introduction to PSL , 2006, Series on Integrated Circuits and Systems.

[15]  Sanjit A. Seshia,et al.  Mining assumptions for synthesis , 2011, Ninth ACM/IEEE International Conference on Formal Methods and Models for Codesign (MEMPCODE2011).

[16]  Markus Holzer,et al.  From Finite Automata to Regular Expressions and Back - A Summary on Descriptional Complexity , 2014, Int. J. Found. Comput. Sci..

[17]  Evelina Lamma,et al.  Inducing Declarative Logic-Based Models from Labeled Traces , 2007, BPM.

[18]  Andreas Zeller,et al.  Mining temporal specifications from object usage , 2011, Automated Software Engineering.

[19]  Dana Fisman,et al.  Functional Specification of Hardware via Temporal Logic , 2018, Handbook of Model Checking.

[20]  Evelina Lamma,et al.  Exploiting Inductive Logic Programming Techniques for Declarative Process Mining , 2009, Trans. Petri Nets Other Model. Concurr..

[21]  Daniel Neider Computing Minimal Separating DFAs and Regular Invariants Using SAT and SMT Solvers , 2012, ATVA.

[22]  Marijn J. H. Heule,et al.  SAT Competition 2016: Recent Developments , 2017, AAAI.

[23]  Calin Belta,et al.  Temporal Logics for Learning and Detection of Anomalous Behavior , 2017, IEEE Transactions on Automatic Control.