Bio-inspired meta-learning for active exploration during non-stationary multi-armed bandit tasks

@article{Velentzas2017BioinspiredMF,
  title={Bio-inspired meta-learning for active exploration during non-stationary multi-armed bandit tasks},
  author={George Velentzas and Costas Tzafestas and Mehdi Khamassi},
  journal={2017 Intelligent Systems Conference (IntelliSys)},
  year={2017},
  pages={661-669}
}
Fast adaptation to changes in the environment requires agents (animals, robots and simulated artefacts) to be able to dynamically tune an exploration-exploitation trade-off during learning. This trade-off usually determines a fixed proportion of exploitative choices (i.e. choice of the action that subjectively appears as best at a given moment) relative to exploratory choices (i.e. testing other actions that now appear worst but may turn out promising later). Rather than using a fixed… CONTINUE READING

References

Publications referenced by this paper.
SHOWING 1-10 OF 22 REFERENCES

Discounted ucb

L. Kocsis, C. Szepesvári
  • 2nd PASCAL Challenges Workshop, 2006, pp. 784–791.
  • 2006
VIEW 16 EXCERPTS
HIGHLY INFLUENTIAL

Meta-learning in Reinforcement Learning

  • Neural Networks
  • 2003
VIEW 5 EXCERPTS
HIGHLY INFLUENTIAL

Diversity-driven selection of exploration strategies in multi-armed bandits

  • 2015 Joint IEEE International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob)
  • 2015
VIEW 1 EXCERPT

EXP3 with drift detection for the switching bandit problem

  • 2015 IEEE International Conference on Data Science and Advanced Analytics (DSAA)
  • 2015
VIEW 1 EXCERPT

Fraud , “ Exp 3 with drift detection for the switching bandit problem

R. Allesiardo, R.
  • Data Science and Advanced Analytics ( DSAA )
  • 2015

Information processing in decision-making systems

M. van der Meer, Z. Kurth-Nelson, A. D. Redish
  • The Neuroscientist, vol. 18, no. 4, pp. 342–359, 2012.
  • 2012
VIEW 1 EXCERPT

Similar Papers

Loading similar papers…