Sequential Decision Making Using Q Learning Algorithm for Diabetic Patients

Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 324)

Abstract

In sequential decision making, we program agent by reward and punishment. In this, agent learns to map situations to actions which results in maximizing rewards gained. This agent is also known as decision makers. It is difficult to take decision about giving specific kind and quantity of insulin dose to the diabetes patient in a critical system of insulin pump control. This paper implements the Q learning algorithm on diabetes data streams. This helps in classifying the data for diabetes dose and also helps in making decision about giving particular kind and quantity of insulin dose by generating various rules.

Keywords

Decision making Diabetes Reinforcement learning Q learning algorithms 

References

  1. 1.
    L.P. Kaelbling et al., in Reinforcement Learning: A Survey. AI Access Foundation and Morgan Kaufmann Publishers. All rights reserved (1996)Google Scholar
  2. 2.
    M.A. Wiering, H. van Hasselt, Ensemble algorithms in reinforcement learning. IEEE Trans. Syst., Man, Cybern.—Part b: Cybern. 38(4)(2008)Google Scholar
  3. 3.
    M. Mill´an-Giraldo, et al., in On-line Classification of Data Streams with Missing Values based on Reinforcement Learning. Institute of New Imaging TechnologiesGoogle Scholar
  4. 4.
    M.M. Gaber, A. Zaslavsky, S. Krishnaswamy. Mining data streams: a review. SIGMOD Rec. 34(2) (2005)Google Scholar
  5. 5.
    R. Garnett, S.J. Roberts, in Learning from Data Streams with Concept Drift. Technical Report PARG-08-01, Hilary Term (2008)Google Scholar
  6. 6.
    L. Busoniu et al., A comprehensive survey of multiagent reinforcement learning. IEEE Trans. Syst., Man, Cybern.—Part c: Appl. Rev. 38(2) (2008)Google Scholar
  7. 7.
    C. Gaskett, D. Wettergreen, A. Zelinsky, Q-learning in continuous state and action spaces. Int. J. Comput. Sci. Eng. (IJCSE)Google Scholar
  8. 8.
    H. van Seijen, H. van Hasselt, S. Whiteson, M. Wiering. A Theoretical and Empirical Analysis of Expected Sarsa Google Scholar
  9. 9.
    M.A. Wiering, QV (lambda)-learning: a new on-policy reinforcement learning algorithm, in Proceedings of the 7th European Workshop on Reinforcement Learning (2005) pp. 29–30Google Scholar
  10. 10.
    A.G. Barto et al. in Learning and Sequential Decision Making. pp. 530–599Google Scholar
  11. 11.
    G. Ditzler, R. Polikar, Incremental learning of concept drift from streaming imbalanced data. IEEE Trans. Know. Data Eng. 25, 2283–2301 (2013)Google Scholar
  12. 12.
    C. Gaskett, D. Wettergreen, A. Zelinsky, Q-learning in continuous state and action spaces . Int. J. Comput. Sci. Eng. (IJCSE)Google Scholar
  13. 13.
    S. Harm van Seijen, H. van Hasselt, S. Whiteson, M. Wiering. A theoretical and empirical analysis of expectedGoogle Scholar

Copyright information

© Springer India 2015

Authors and Affiliations

  • Pramod Patil
    • 1
  • Parag Kulkarni
    • 1
  • Rachana Shirsath
    • 2
  1. 1.College of Engineering PunePuneIndia
  2. 2.Dr. D.Y. Patil Institute of Engineering and TechnologyPuneIndia

Personalised recommendations