Learning to Evaluate Go Positions via Temporal Difference Methods

  title={Learning to Evaluate Go Positions via Temporal Difference Methods},
  author={Nicol N. Schraudolph and Peter Dayan and Terrence J. Sejnowski},
The game of Go has a high branching factor that defeats the tree search approach used in computer chess, and long-range spatiotemporal interactions that make position evaluation extremely difficult. Development of conventional Go programs is hampered by their knowledge-intensive nature. We demonstrate a viable alternative by training neural networks to evaluate Go positions via temporal difference (TD) learning. 
Highly Cited
This paper has 43 citations. REVIEW CITATIONS


Publications citing this paper.
Showing 1-10 of 24 extracted citations


Publications referenced by this paper.
Showing 1-10 of 20 references

To test a powerful computer, play an ancient game

  • G. Johnson
  • The New York Times, July 29,
  • 1997
1 Excerpt

The integration of a priori knowledge into a Go playing neural network

  • M. Enzensberger
  • 1996
1 Excerpt

Monte Carlo Go

  • B. Brügmann
  • 1993
2 Excerpts

Similar Papers

Loading similar papers…