Abstract
Exploiting symmetries is an important topic to obtain sparse (lifted) representations, reduce complexity and achieve good performance in dynamic probabilistic relational models (DPRMs). DPRMs factorise a full joint probability distribution by encoding multivariate time series through a set of conditionally dependent random variables. As obtaining exact symmetries throughout multivariate time series is often not realistic in real-world contexts and counteracts lifted representations, we propose to approximate the multivariate time series with a symbolisation scheme that encodes the overarching trend in up and down movements. In this work, we introduce MOP4SA, an approach for the approximation of symmetries based on multivariate ordinal pattern encodings and spectral clustering. Understanding symmetrical behaviour has several benefits that we evaluate in two use cases. We use MOP4SA (a) to detect structures in model symmetries over time, and (b) to avoid model splits and groundings in DPRMs.
N. Finke and M. Mohr—Contributed equally to this work.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Pronounced deeper models.
- 2.
- 3.
References
Agostini, A., Celaya, E.: Exploiting domain symmetries in reinforcement learning with continuous state and action spaces. In: 2009 International Conference on Machine Learning and Applications, Miami, FL, USA, pp. 331–336. IEEE (2009)
Agrawal, R., Faloutsos, C., Swami, A.: Efficient similarity search in sequence databases. In: Lomet, D.B. (ed.) FODO 1993. LNCS, vol. 730, pp. 69–84. Springer, Heidelberg (1993). https://doi.org/10.1007/3-540-57301-1_5
Bandt, C., Pompe, B.: Permutation entropy: a natural complexity measure for time series. Phys. Rev. Lett. 88(17), 4 (2002)
Bellman, R.: Adaptive Control Processes: A guided Tour. Princeton Legacy Library, Princeton University Press, Princeton (2015)
Bertozzi, A.L., Merkurjev, E.: Chapter 12 - Graph-based optimization approaches for machine learning, uncertainty quantification and networks. In: Kimmel, R., Tai, X.C. (eds.) Processing, Analyzing and Learning of Images, Shapes, and Forms: Part 2, Handbook of Numerical Analysis, vol. 20, pp. 503–531. Elsevier (2019)
Chiu, B., Keogh, E., Lonardi, S.: Probabilistic discovery of time series motifs. In: Proceedings of the Ninth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 493–498 (2003)
Dieleman, S., Fauw, J.D., Kavukcuoglu, K.: Exploiting cyclic symmetry in convolutional neural networks. In: Proceedings of the 33rd International Conference on Machine Learning, pp. 1889–1898 (2016)
Finke, N., Gehrke, M., Braun, T., Potten, T., Möller, R.: Investigating matureness of probabilistic graphical models for dry-bulk shipping. In: Proceedings of the 10th International Conference on Probabilistic Graphical Models, pp. 197–208 (2020)
Finke, N., Mohr, M.: A priori approximation of symmetries in dynamic probabilistic relational models. In: Edelkamp, S., Möller, R., Rueckert, E. (eds.) KI 2021. LNCS (LNAI), vol. 12873, pp. 309–323. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-87626-5_23
Gehrke, M., Braun, T., Möller, R.: Lifted dynamic junction tree algorithm. In: Chapman, P., Endres, D., Pernelle, N. (eds.) ICCS 2018. LNCS (LNAI), vol. 10872, pp. 55–69. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-91379-7_5
Keogh, E., Chakrabarti, K., Pazzani, M., Mehrotra, S.: Dimensionality reduction for fast similarity search in large time series databases. In: Knowledge and Information Systems, pp. 263–286 (2001). https://doi.org/10.1021/acsami.7b03579
Kramer, S.: A brief history of learning symbolic higher-level representations from data (and a curious look forward). In: Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence, IJCAI-20, pp. 4868–4876 (2020)
Kruskal, J.B., Liberman, M.: The symmetric time warping problem: From continuous to discrete. In: Time Warps, String Edits and Macromolecules: The Theory and Practice of Sequence Comparison. Addison-Wesley Publishing Co. (1983)
Mohr, M., Wilhelm, F., Hartwig, M., Möller, R., Keller, K.: New approaches in ordinal pattern representations for multivariate time series. In: Proceedings of the 33rd International Florida Artificial Intelligence Research Society Conference (2020)
Myers, A., Khasawneh, F.A.: On the automatic parameter selection for permutation entropy. Chaos: an Interdis. J. Nonlinear Sci. 30(3), 033130 (2020)
Niepert, M., Van den Broeck, G.: Tractability through exchangeability: a new perspective on efficient probabilistic inference. In: AAAI-14 Proceedings of the 28th AAAI Conference on Artificial Intelligence, pp. 2467–2475. AAAI Press (2014)
Petitjean, F., Inglada, J., Gancarski, P.: Satellite image time series analysis under time warping. IEEE Trans. Geosci. Remote Sens. 50(8) (2012)
Poole, D.: First-order probabilistic inference. In: Proceedings of the 18th International Joint Conference on Artificial Intelligence, pp. 985–991 (2003)
Salvador, S., Chan, P.: FastDTW: toward accurate dynamic time warping in linear time and space. In: KDD Workshop on Mining Temporal and Sequential Data, pp. 70–80 (2004)
Sato, T.: A Statistical learning method for logic programs with distribution semantics. In: Proceedings of the 12th International Conference on Logic Programming, pp. 715–729 (1995)
Satorras, V.G., Hoogeboom, E., Welling, M.: E(n) equivariant graph neural networks. In: Proceedings of the 38th ICML (2021)
Yeh, C.C.M., et al.: Matrix profile I: all pairs similarity joins for time series: a unifying view that includes motifs, discords and shapelets. In: 2016 IEEE 16th International Conference on Data Mining (ICDM), pp. 1317–1322 (2016)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Finke, N., Möller, R., Mohr, M. (2022). Multivariate Ordinal Patterns for Symmetry Approximation in Dynamic Probabilistic Relational Models. In: Long, G., Yu, X., Wang, S. (eds) AI 2021: Advances in Artificial Intelligence. AI 2022. Lecture Notes in Computer Science(), vol 13151. Springer, Cham. https://doi.org/10.1007/978-3-030-97546-3_44
Download citation
DOI: https://doi.org/10.1007/978-3-030-97546-3_44
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-97545-6
Online ISBN: 978-3-030-97546-3
eBook Packages: Computer ScienceComputer Science (R0)