Schemes for learning and behaviour : a new expectancy model

@inproceedings{Witkowski1997SchemesFL,
  title={Schemes for learning and behaviour : a new expectancy model},
  author={Christopher Mark Witkowski},
  year={1997}
}
This thesis presents a novel form of learning by reinforcement. Existing reinforcement learning algorithms rely on the provision of external reward signals to drive the learning algorithm. This new algorithm relies on reinforcing signals generated internally within the algorithm. The algorithm, SRS/E, described here generates expectancies ( -hypotheses), each of which gives rise to a specific prediction when the conditions relevant to the expectancy are encountered (the -experiment). The… CONTINUE READING
Highly Cited
This paper has 33 citations. REVIEW CITATIONS

Citations

Publications citing this paper.

References

Publications referenced by this paper.
Showing 1-10 of 25 references

134 4-8: Step One

  • Srs E Algorithm The
  • Step Two, Evaluation of -Experiments…

157 6-1: Results from Sutton's Dyna-PI

  • Sutton 's Dynaworld, Environment
  • The SRS/E Experimenter Command Options…

251 List of Figures and Tables 1-1: Tinbergen's Principle of

  • References
  • SUBJECT and AUTHOR INDEX…

54 2-7: The Marginal Attribution

  • Composite Action
  • Process…

A Simple Neurone Model ..................................................................................... A Multilayer Neural Network Model

  • Classifier System
  • A Simple Neurone Model…

Action Selection Architecture

  • Action Selection Architecture

An Association List

  • An Association List

An Explicit Template List

  • An Explicit Template List

Appendix One

  • Appendix One

Aversion

  • Aversion

Similar Papers

Loading similar papers…