Dana Nau, Paul Purdom, and Chun-Hung Tzeng. An evaluation of two alternatives to minimax. In L. N. Kanal and J. F. Lemmer, editors, Uncertainty in Artificial Intelligence, pages 505--509. Elsevier Science Publishers B.V., The Netherlands, 1988.  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... to find the results of the optimal strategy for the subtree (therefore producing exact values) Thus, the solution to improving the double dummy architectures of Levy and Ginsberg does not lie with the simple replacement of the minimax algorithm with modified versions, such as average propagation [Nau et al. 88] or product propagation CHAPTER 6. SEARCH IN GAMES WITH INCOMPLETE INFORMATION 141 [Pearl 81, Pearl 83] The back up rules in such algorithms, when calculating a value to propagate up the tree, take into account the value of each subtree at a node. Their use in the repeated minimaxing algorithm ....

Dana Nau, Paul Purdom, and Chun-Hung Tzeng. An evaluation of two alternatives to minimax. In L. N. Kanal and J. F. Lemmer, editors, Uncertainty in Artificial Intelligence, pages 505--509. Elsevier Science Publishers B.V., The Netherlands, 1988.

.... to find the results of the optimal strategy for the subtree (therefore producing exact values) Thus, the solution to improving the double dummy architectures of Levy and Ginsberg does not lie with the simple replacement of the minimax algorithm with modified versions, such as average propagation [Nau et al. 88] or product propagation [Pearl 81, Pearl 83] The back up rules in such algorithms, when calculating a value to propagate up the tree, take into account the value of each subtree at a node. Their use in the repeated minimaxing algorithm may therefore reduce the effects of strategy mixing, but ....

Dana Nau, Paul Purdom, and Chun-Hung Tzeng. An evaluation of two alternatives to minimax. In L. N. Kanal and J. F. Lemmer, editors, Uncertainty in Artificial Intelligence, pages 505--509. Elsevier Science Publishers B.V., The Netherlands, 1988.

.... by strategy fusion) or by applying exhaustive strategy minimisation to find the results of the optimal strategy for the subtree (therefore producing exact values) Thus, non locality cannot be eradicated by simply replacing the minimax algorithm with modified versions, such as average propagation [26] or product propagation [28,29] The back up rules in such algorithms, when calculating a value to propagate up the tree, take into account the value of each subtree at a node. Their use in the repeated minimaxing architecture may therefore reduce the effects of strategy fusion, but non locality ....

Dana Nau, Paul Purdom, and Chun-Hung Tzeng. An evaluation of two alternatives to minimax. In L. N. Kanal and J. F. Lemmer, editors, Uncertainty in Artificial Intelligence, pages 505--509. Elsevier Science Publishers B.V., The Netherlands, 1988.

....0 and 1. However, the assumption of independence of the evaluation of sibling nodes is generally attacked as being unrealistic. Unfortunately, this backup rule is not useful if some of the successors are evaluated with bounds (as alpha beta does) 2.1. 3 Average Propagation Nau, Purdon and Tzeng [NPH86] argue that since both minimax propagation and product propagation have their justification in actual game playing and search models, an average of both propagations could be useful. 2 ALTERNATIVE APPROACHES 5 Again, the backup procedure is just part of an algorithm. The proposal is to ....

D.S. Nau, P. Purdom, and Tzeng H.C. An evaluation on two alternatives to minimax. In Uncertainty in Artificial Intelligence, pages 505--509, North Holland, Amsterdam, 1986.

....careful organization and allocation of computational resources is needed to obtain expert level play. Techniques such as alpha beta pruning and its successors [6, 2] have been essential in reducing the computational burden of exploring a game tree. Still, new techniques are needed. Nau et al. [10], after much expermentation with existing methods, assert that A method is needed which will always expand the node that is expected to have the largest effect on the value . This paper suggests such a method. Our method, min max approximation , attempts to focus the computer s attention on the ....

Nau, Dana S., Paul Purdom, and Chun-Hung Tzeng, "An evaluation of two alternatives to minimax," Proc. on Uncertainty and Probability in Artificial Intelligence (AAAI, UCLA, August 14--16, 1985), 232--235.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC