Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
149 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

A Systems Theoretic Approach to Online Machine Learning (2404.03775v1)

Published 4 Apr 2024 in cs.LG, cs.AI, cs.SY, and eess.SY

Abstract: The machine learning formulation of online learning is incomplete from a systems theoretic perspective. Typically, machine learning research emphasizes domains and tasks, and a problem solving worldview. It focuses on algorithm parameters, features, and samples, and neglects the perspective offered by considering system structure and system behavior or dynamics. Online learning is an active field of research and has been widely explored in terms of statistical theory and computational algorithms, however, in general, the literature still lacks formal system theoretical frameworks for modeling online learning systems and resolving systems-related concept drift issues. Furthermore, while the machine learning formulation serves to classify methods and literature, the systems theoretic formulation presented herein serves to provide a framework for the top-down design of online learning systems, including a novel definition of online learning and the identification of key design parameters. The framework is formulated in terms of input-output systems and is further divided into system structure and system behavior. Concept drift is a critical challenge faced in online learning, and this work formally approaches it as part of the system behavior characteristics. Healthcare provider fraud detection using machine learning is used as a case study throughout the paper to ground the discussion in a real-world online learning challenge.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (40)
  1. P. Zhao, S. C. Hoi, J. Wang, and B. Li, “Online transfer learning,” Artificial intelligence, vol. 216, pp. 76–102, 2014.
  2. S. C. Hoi, D. Sahoo, J. Lu, and P. Zhao, “Online learning: A comprehensive survey,” Neurocomputing, vol. 459, pp. 249–289, 2021.
  3. H. B. McMahan, “A survey of algorithms and analysis for adaptive online learning,” The Journal of Machine Learning Research, vol. 18, no. 1, pp. 3117–3166, 2017.
  4. J. Lu, A. Liu, F. Dong, F. Gu, J. Gama, and G. Zhang, “Learning under concept drift: A review,” IEEE transactions on knowledge and data engineering, vol. 31, no. 12, pp. 2346–2363, 2018.
  5. J. Gama, I. Žliobaitė, A. Bifet, M. Pechenizkiy, and A. Bouchachia, “A survey on concept drift adaptation,” ACM computing surveys (CSUR), vol. 46, no. 4, pp. 1–37, 2014.
  6. S. Shalev-Shwartz et al., “Online learning and online convex optimization,” Foundations and Trends® in Machine Learning, vol. 4, no. 2, pp. 107–194, 2012.
  7. N. Cesa-Bianchi and F. Orabona, “Online learning algorithms,” Annual review of statistics and its application, vol. 8, pp. 165–190, 2021.
  8. D. Dori, H. Sillitto, R. M. Griego, D. McKinney, E. P. Arnold, P. Godfrey, J. Martin, S. Jackson, and D. Krob, “System definition, system worldviews, and systemness characteristics,” IEEE Systems Journal, vol. 14, no. 2, pp. 1538–1548, 2019.
  9. M. D. Mesarovic and Y. Takahara, General systems theory: mathematical foundations. Academic press, 1975.
  10. M. D. Mesarovic and Y. Takahara, “Abstract systems theory,” 1989.
  11. T. Cody and P. A. Beling, “A systems theory of transfer learning,” IEEE Systems Journal, vol. 17, no. 1, pp. 26–37, 2023.
  12. T. Cody, “Homomorphisms between transfer, multi-task, and meta-learning systems,” in International Conference on Artificial General Intelligence, pp. 199–208, Springer, 2022.
  13. T. Cody, “Mesarovician abstract learning systems,” in Artificial General Intelligence: 14th International Conference, AGI 2021, Palo Alto, CA, USA, October 15–18, 2021, Proceedings 14, pp. 55–64, Springer, 2022.
  14. X.-C. Yin, K. Huang, and H.-W. Hao, “Dynamic ensemble of ensembles in nonstationary environments,” in Neural Information Processing: 20th International Conference, ICONIP 2013, Daegu, Korea, November 3-7, 2013. Proceedings, Part II 20, pp. 76–83, Springer, 2013.
  15. “The national heath care anti-fraud association: The challenge of health care fraud.”
  16. T. Cody, S. Adams, and P. Beling, “Motivating a systems theory of ai,” Insight, vol. 23, no. 1, pp. 37–40, 2020.
  17. T. Cody and P. A. Beling, “Cascading negative transfer in networks of machine learning systems,” in 2023 IEEE International Conference on Assured Autonomy (ICAA), pp. 141–148, IEEE, 2023.
  18. J. Z. Kolter and M. A. Maloof, “Dynamic weighted majority: An ensemble method for drifting concepts,” The Journal of Machine Learning Research, vol. 8, pp. 2755–2790, 2007.
  19. G. Jaber, A. Cornuéjols, and P. Tarroux, “Online learning: Searching for the best forgetting strategy under concept drift,” in Neural Information Processing: 20th International Conference, ICONIP 2013, Daegu, Korea, November 3-7, 2013. Proceedings, Part II 20, pp. 400–408, Springer, 2013.
  20. F. Hinder, V. Vaquet, J. Brinkrolf, and B. Hammer, “On the hardness and necessity of supervised concept drift detection,” 2023.
  21. K. Goel and S. Batra, “Adaptive online learning for classification under concept drift,” International Journal of Computational Science and Engineering, vol. 24, no. 2, pp. 128–135, 2021.
  22. S.-C. You and H.-T. Lin, “A simple unlearning framework for online learning under concept drifts,” in Advances in Knowledge Discovery and Data Mining: 20th Pacific-Asia Conference, PAKDD 2016, Auckland, New Zealand, April 19-22, 2016, Proceedings, Part I 20, pp. 115–126, Springer, 2016.
  23. V. M. Patel, R. Gopalan, R. Li, and R. Chellappa, “Visual domain adaptation: A survey of recent advances,” IEEE signal processing magazine, vol. 32, no. 3, pp. 53–69, 2015.
  24. S. Ben-David, J. Blitzer, K. Crammer, and F. Pereira, “Analysis of representations for domain adaptation,” Advances in neural information processing systems, vol. 19, 2006.
  25. A. Tsymbal, “The problem of concept drift: definitions and related work,” Computer Science Department, Trinity College Dublin, vol. 106, no. 2, p. 58, 2004.
  26. S. M. Jameel, M. A. Hashmani, H. Alhussain, M. Rehman, and A. Budiman, “A critical review on adverse effects of concept drift over machine learning classification models,” International Journal of Advanced Computer Science and Applications, vol. 11, no. 1, 2020.
  27. I. Goldenberg and G. I. Webb, “Survey of distance measures for quantifying concept drift and shift in numeric data,” Knowledge and Information Systems, vol. 60, no. 2, pp. 591–615, 2019.
  28. R. Vilalta and Y. Drissi, “A perspective view and survey of meta-learning,” Artificial intelligence review, vol. 18, pp. 77–95, 2002.
  29. R. Vilalta, C. Giraud-Carrier, and P. Brazdil, “Meta-learning-concepts and techniques,” Data mining and knowledge discovery handbook, pp. 717–731, 2010.
  30. K. A. Smith-Miles, “Cross-disciplinary perspectives on meta-learning for algorithm selection,” ACM Computing Surveys (CSUR), vol. 41, no. 1, pp. 1–25, 2009.
  31. D. Li, Y. Yang, Y.-Z. Song, and T. Hospedales, “Learning to generalize: Meta-learning for domain generalization,” in Proceedings of the AAAI conference on artificial intelligence, vol. 32, 2018.
  32. F. Hinder, A. Artelt, and B. Hammer, “Towards non-parametric drift detection via dynamic adapting window independence drift detection (dawidd),” in International Conference on Machine Learning, pp. 4249–4259, PMLR, 2020.
  33. J. Z. Kolter and M. A. Maloof, “Using additive expert ensembles to cope with concept drift,” in Proceedings of the 22nd international conference on Machine learning, pp. 449–456, 2005.
  34. H. Wu, Y. Yan, Y. Ye, H. Min, M. K. Ng, and Q. Wu, “Online heterogeneous transfer learning by knowledge transition,” ACM Transactions on Intelligent Systems and Technology (TIST), vol. 10, no. 3, pp. 1–19, 2019.
  35. Q. Wu, H. Wu, X. Zhou, M. Tan, Y. Xu, Y. Yan, and T. Hao, “Online transfer learning with multiple homogeneous or heterogeneous sources,” IEEE Transactions on Knowledge and Data Engineering, vol. 29, no. 7, pp. 1494–1507, 2017.
  36. P. ZHAO and S. C. HOI, “Otl: A framework of online transfer learning.(2010),” in Proceedings of the 27th International Conference on Machine Learning (ICML 2010): Haifa, Israel, 21-24 June, pp. 219–1.
  37. L. Ge, J. Gao, and A. Zhang, “Oms-tl: A framework of online multiple source transfer learning,” in Proceedings of the 22nd ACM international conference on Information & Knowledge Management, pp. 2423–2428, 2013.
  38. T. Grubinger, G. C. Chasparis, and T. Natschläger, “Generalized online transfer learning for climate control in residential buildings,” Energy and Buildings, vol. 139, pp. 63–71, 2017.
  39. T. Cody, S. Adams, and P. A. Beling, “A systems theoretic perspective on transfer learning,” in 2019 IEEE International Systems Conference (SysCon), pp. 1–7, IEEE, 2019.
  40. A. Budiman, M. I. Fanany, C. Basaruddin, et al., “Adaptive online sequential elm for concept drift tackling,” Computational intelligence and neuroscience, vol. 2016, 2016.
Citations (2)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets