ABSTRACT
Many real world stochastic processes are non-stationary, which means that the probability distribution that generates data samples is time-varying. In the context of machine learning, this phenomenon is known as concept drift. It is important that machine learning models are able to adapt to concept drift in order to prevent degradation in accuracy. In this paper, we present two algorithms for drift detection and adaptation.
Drift is measured by continuously tracking a difference metric between probability distributions estimated from two sample windows preceding a time point. High values for the difference metric indicates that concept drift has occurred, and the model must be adapted. Adaptation is done by training a new model for the drifted process, and adding it to an ensemble of models. Previously trained models are retained, and their weights in the ensemble are adjusted to reflect similarity with the current probability distribution of the process. Experiments on simulated drift scenarios as well as real world datasets show that our algorithms detect drift with high accuracy, and adaptation results in improved model accuracy.
- Webb, Geoffrey I., et al. "Understanding Concept Drift." arXiv preprint arXiv:1704.00362 (2017).Google Scholar
- Gama, Joao, et al. "Learning with drift detection." Brazilian Symposium on Artificial Intelligence. Springer, Berlin, Heidelberg, 2004.Google Scholar
- Žliobaitė, Indrė. "Learning under concept drift: an overview." arXiv preprint arXiv:1010.4784 (2010).Google Scholar
- Tsymbal, Alexey. "The problem of concept drift: definitions and related work." Computer Science Department, Trinity College Dublin 106.2 (2004).Google Scholar
- Street, W. Nick, and YongSeog Kim. "A streaming ensemble algorithm (SEA) for large-scale classification." Proceedings of the seventh ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, 2001. Google ScholarDigital Library
- Wang, Haixun, et al. "Mining concept-drifting data streams using ensemble classifiers." Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining. AcM, 2003. Google ScholarDigital Library
- I. Koychev. Koychev, Ivan. "Gradual forgetting for adaptation to concept drift." Proceedings of ECAI 2000 Workshop on Current Issues in Spatio-Temporal Reasoning, 2000.Google Scholar
- Zhang, Peng, Xingquan Zhu, and Yong Shi. "Categorizing and mining concept drifting data streams." Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, 2008. Google ScholarDigital Library
- Scholz, Martin, and Ralf Klinkenberg. "An ensemble classifier for drifting concepts." Proceedings of the Second International Workshop on Knowledge Discovery in Data Streams. Porto, Portugal, 2005.Google Scholar
- Royer, Amelie, and Christoph H. Lampert. "Classifier adaptation at prediction time." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2015.Google Scholar
- Hoffman, Judy, Trevor Darrell, and Kate Saenko. "Continuous manifold based adaptation for evolving visual domains." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2014. Google ScholarDigital Library
- Levinkov, Evgeny, and Mario Fritz. "Sequential Bayesian model update under structured scene prior for semantic road scenes labeling." Proceedings of the IEEE International Conference on Computer Vision. 2013. Google ScholarDigital Library
- Levin, David Asher, Yuval Peres, and Elizabeth Lee Wilmer. Markov chains and mixing times. American Mathematical Soc., 2009.Google Scholar
- Michael Harries. Splice-2 comparative evaluation: Electricity pricing. Technical reportGoogle Scholar
- Bifet, Albert, and Ricard Gavalda. "Learning from time-changing data with adaptive windowing." Proceedings of the 2007 SIAM International Conference on Data Mining. Society for Industrial and Applied Mathematics, 2007.Google Scholar
- Raykar, Vikas C., Ramani Duraiswami, and Linda H. Zhao. "Fast computation of kernel estimators." Journal of Computational and Graphical Statistics 19.1 (2010): 205--220.Google ScholarCross Ref
- Elgammal, Ahmed, Ramani Duraiswami, and Larry S. Davis. "Efficient kernel density estimation using the fast gauss transform with applications to color modeling and tracking." IEEE transactions on pattern analysis and machine intelligence 25.11 (2003): 1499--1504. Google ScholarDigital Library
Index Terms
- Detecting and Adapting to Concept Drift in Continually Evolving Stochastic Processes
Recommendations
Unsupervised Concept Drift Detection with a Discriminative Classifier
CIKM '19: Proceedings of the 28th ACM International Conference on Information and Knowledge ManagementIn data stream mining, one of the biggest challenges is to develop algorithms that deal with the changing data. As data evolve over time, static models become outdated. This phenomenon is called concept drift, and it is investigated extensively in the ...
Concept Drift Adaptation by Exploiting Drift Type
Concept drift is a phenomenon where the distribution of data streams changes over time. When this happens, model predictions become less accurate. Hence, models built in the past need to be re-learned for the current data. Two design questions need to be ...
Brute force concept drift detection
AbstractWe present a brute-force approach to detect concept drift behind time sequence data. This approach, named Select-Starţ searches for start points of concept drift to minimize error. In other words, Select-Start searches for the start points of new ...
Comments