Papers
Topics
Authors
Recent
Search
2000 character limit reached

Learning temporal formulas from examples is hard

Published 26 Dec 2023 in cs.LG, cs.AI, cs.FL, and cs.LO | (2312.16336v1)

Abstract: We study the problem of learning linear temporal logic (LTL) formulas from examples, as a first step towards expressing a property separating positive and negative instances in a way that is comprehensible for humans. In this paper we initiate the study of the computational complexity of the problem. Our main results are hardness results: we show that the LTL learning problem is NP-complete, both for the full logic and for almost all of its fragments. This motivates the search for efficient heuristics, and highlights the complexity of expressing separating properties in concise natural language.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (17)
  1. Uncertainty-aware signal temporal logic inference.Ā  In Bloem, R., Dimitrova, R., Fan, C., &Ā Sharygina, N. (Eds.), International Conference on Software Verification, VSTTE, Vol.Ā 13124 of Lecture Notes in Computer Science, pp.Ā 61–85. Springer.
  2. A Decision Tree Approach to Data Classification using Signal Temporal Logic.Ā  In Hybrid Systems: Computation and Control, HSCC.
  3. Learning interpretable models expressed in linear temporal logic.Ā  International Conference on Automated Planning and Scheduling, ICAPS, 29.
  4. Analytical approach to parallel repetition.Ā  In Symposium on Theory of Computing, STOC, pp.Ā 624–633.
  5. Learning properties in LTL ∩\cap∩ ACTL from positive examples only.  In Formal Methods in Computer Aided Design, FMCAD.
  6. The complexity of learning linear temporal formulas from examples.Ā  In Chandlee, J., Eyraud, R., Heinz, J., Jardine, A., &Ā van Zaanen, M. (Eds.), International Conference on Grammatical Inference, ICGI, Vol.Ā 153 of Proceedings of Machine Learning Research, pp.Ā 237–250. PMLR.
  7. Maxsat-based temporal logic inference from noisy data.Ā  Innovations in Systems and Software Engineering, 18(3), 427–442.
  8. Gold, E.Ā M. (1978). Complexity of automaton identification from given data.Ā  Information and Control, 37(3), 302–320.
  9. Bayesian inference of linear temporal logic specifications for contrastive explanations.Ā  In International Joint Conference on Artificial Intelligence, IJCAI.
  10. General LTL specification mining.Ā  In International Conference on Automated Software Engineering, (ASE).
  11. Specification sketching for linear temporal logic.Ā  In AndrĆ©, Ɖ.,Ā  &Ā Sun, J. (Eds.), International Symposium on Automated Technology for Verification and Analysis, ATVA, Vol.Ā 14216 of Lecture Notes in Computer Science, pp.Ā 26–48. Springer.
  12. Learning linear temporal properties.Ā  In Formal Methods in Computer Aided Design, FMCAD, pp.Ā 1–10.
  13. The minimum consistent DFA problem cannot be approximated within any polynomial.Ā  Journal of the ACM, 40(1), 95–142.
  14. Pnueli, A. (1977). The temporal logic of programs.Ā  In Symposium on Foundations of Computer Science, SFCS.
  15. Scalable anytime algorithms for learning fragments of linear temporal logic.Ā  In Fisman, D.,Ā  &Ā Rosu, G. (Eds.), International Conference on Tools and Algorithms for the Construction and Analysis of Systems, TACAS, Vol.Ā 13243 of Lecture Notes in Computer Science, pp.Ā 263–280. Springer.
  16. Synthesizing efficiently monitorable formulas in metric temporal logic.Ā  In International Conference on Verification, Model Checking, and Abstract Interpretation, VMCAI.
  17. Learning interpretable models in the property specification language.Ā  In International Joint Conference on Artificial Intelligence, IJCAI.
Citations (2)

Summary

No one has generated a summary of this paper yet.

Paper to Video (Beta)

No one has generated a video about this paper yet.

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Continue Learning

We haven't generated follow-up questions for this paper yet.

Collections

Sign up for free to add this paper to one or more collections.