A Problem Case for UCT

@article{Browne2013APC,
  title={A Problem Case for UCT},
  author={Cameron Browne},
  journal={IEEE Transactions on Computational Intelligence and AI in Games},
  year={2013},
  volume={5},
  pages={69-74}
}
This paper examines a simple 5 × 5 Hex position that not only completely defeats flat Monte Carlo search, but also initially defeats plain upper confidence bounds for trees (UCT) search until an excessive number of iterations are performed. The inclusion of domain knowledge during playouts significantly improves UCT performance, but a slight negative effect is shown for the rapid action value estimate (RAVE) heuristic under some circumstances. This example was drawn from an actual game during… CONTINUE READING

Figures, Tables, and Topics from this paper.

Citations

Publications citing this paper.

References

Publications referenced by this paper.
SHOWING 1-10 OF 17 REFERENCES

On the dangers of random playouts

  • C. Browne
  • Int. Comput. Games Assoc. J., vol. 34, no. 1, pp…
  • 2011
1 Excerpt

Monte-Carlo search techniques in the modern board game Thurn and Taxis

  • F. C. Schadd
  • M.S. thesis, Dept. Comput. Sci., Maastricht Univ…
  • 2009
1 Excerpt

RAVE problems, Computer-Go mailing list

  • M. Müller
  • [Online]. Available: http://www.mail-archive.com…
  • 2009
2 Excerpts

RAVE problems, Computer-Gomailing list

  • D. Hillis
  • [Online]. Available: http://www.mail-archive.com…
  • 2009
1 Excerpt

Similar Papers

Loading similar papers…