Abstract
Concept drift, i.e., the change of the data generating distribution, can render machine learning models inaccurate. Several works address the phenomenon of concept drift in the streaming context usually assuming that consecutive data points are independent of each other. To generalize to dependent data, many authors link the notion of concept drift to time series. In this work, we show that the temporal dependencies are strongly influencing the sampling process. Thus, the used definitions need major modifications. In particular, we show that the notion of stationarity is not suited for this setup and discuss an alternative we refer to as consistency. We demonstrate that consistency better describes the observable learning behavior in numerical experiments.
Funding in the frame of the ERC Synergy Grant “Water-Futures” No. 951424 is gratefully acknowledged.
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
Similar content being viewed by others
Notes
- 1.
The experimental code as well as all datasets can be found at https://github.com/FabianHinder/Remark-on-Dependent-Drift.
- 2.
Autoregressive models based on Ridge and k-NN regression are considered but only work for Jump. This also rules out supervised detectors that process their losses.
References
Adams, T.M., Nobel, A.B.: Uniform convergence of vapnik-chervonenkis classes under ergodic sampling. Ann. Probab. 38(4), 1345–1367 (2010)
Agarwal, A., Duchi, J.C.: The generalization ability of online algorithms for dependent data. IEEE Trans. Inf. Theory 59(1), 573–587 (2012)
Aminikhanghahi, S., Cook, D.J.: A survey of methods for time series change point detection. Knowl. Inf. Syst. 51(2), 339–367 (2017)
Borodin, A.N.: Stochastic Processes. Springer (2017). https://doi.org/10.1007/978-3-319-62310-8
Dickey, D., Fuller, W.: Distribution of the estimators for autoregressive time series with a unit root. JASA. J. Am. Stat. Assoc. 74, 427–431 (1979)
Ditzler, G., Roveri, M., Alippi, C., Polikar, R.: Learning in nonstationary environments: a survey. IEEE Comp. Int. Mag. 10(4), 12–25 (2015)
Gretton, A., Borgwardt, K.M., Rasch, M.J., Schölkopf, B., Smola, A.J.: A kernel method for the two-sample-problem. In: NIPS, pp. 513–520 (2006)
Hanneke, S., Yang, L.: Statistical learning under nonstationary mixing processes. In: The 22nd International Conference on Artificial Intelligence and Statistics, pp. 1678–1686. PMLR (2019)
Harchaoui, Z., Cappé, O.: Retrospective mutiple change-point estimation with kernels. In: 2007 IEEE/SP 14th Workshop on Statistical Signal Processing, pp. 768–772. IEEE (2007)
Harchaoui, Z., Moulines, E., Bach, F.: Kernel change-point analysis. In: Koller, D., Schuurmans, D., Bengio, Y., Bottou, L. (eds.) NIPS, vol. 21 (2008)
Hinder, F., Artelt, A., Hammer, B.: Towards non-parametric drift detection via dynamic adapting window independence drift detection (DAWIDD). In: International Conference on Machine Learning, pp. 4249–4259. PMLR (2020)
Hinder, F., Vaquet, V., Brinkrolf, J., Hammer, B.: On the change of decision boundary and loss in learning with concept drift. In: Crémilleux, B., Hess, S., Nijssen, S. (eds) International Symposium on Intelligent Data Analysis, pp. 182–194. Springer, Cham (2023). https://doi.org/10.1007/978-3-031-30047-9_15
Hinder, F., Vaquet, V., Brinkrolf, J., Hammer, B.: On the hardness and necessity of supervised concept drift detection. In: 2th International Conference on Pattern Recognition Applications and Methods (2023)
Hinder, F., Vaquet, V., Hammer, B.: Suitability of different metric choices for concept drift detection. In: Bouadi, T., Fromont, E., Hüllermeier, E. (eds.) IDA 2022. LNCS, vol. 13205, pp. 157–170. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-01333-1_13
Hinder, F., Vaquet, V., Hammer, B.: One or two things we know about concept drift – a survey on monitoring evolving environments. arXiv preprint arXiv:2310.15826 (2023)
Hinder, F., Vaquet, V., Hammer, B.: A remark on concept drift for dependent data. arXiv preprint arXiv:2312.10212 (2023)
Kontorovich, L.: Measure Concentration of Strongly Mixing Processes with Applications. Carnegie Mellon University (2007)
Krengel, U.: On the speed of convergence in the ergodic theorem. Monatshefte für Mathematik 86(1), 3–6 (1978)
Kwiatkowski, D., Phillips, P.C., Schmidt, P., Shin, Y.: Testing the null hypothesis of stationarity against the alternative of a unit root: how sure are we that economic time series have a unit root? J. Econometrics 54(1–3), 159–178 (1992)
Lim, B., Zohren, S.: Time-series forecasting with deep learning: a survey. Phil. Trans. R. Soc. A 379(2194), 20200209 (2021)
Lu, J., Liu, A., Dong, F., Gu, F., Gama, J., Zhang, G.: Learning under concept drift: a review. IEEE Trans. Knowl. Data Eng. 31(12), 2346–2363 (2018)
Shalev-Shwartz, S., Ben-David, S.: Understanding Machine Learning: From Theory to Algorithms. Cambridge university press (2014)
Truong, C., Oudre, L., Vayatis, N.: Selective review of offline change point detection methods. Signal Process. 167, 107299 (2020)
Yu, B.: Rates of convergence for empirical processes of stationary mixing sequences. Ann. Probab. 22(1), 94–116 (1994)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Hinder, F., Vaquet, V., Hammer, B. (2024). A Remark on Concept Drift for Dependent Data. In: Miliou, I., Piatkowski, N., Papapetrou, P. (eds) Advances in Intelligent Data Analysis XXII. IDA 2024. Lecture Notes in Computer Science, vol 14641. Springer, Cham. https://doi.org/10.1007/978-3-031-58547-0_7
Download citation
DOI: https://doi.org/10.1007/978-3-031-58547-0_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-58546-3
Online ISBN: 978-3-031-58547-0
eBook Packages: Computer ScienceComputer Science (R0)