A simple easing game solving method based on pruning search

Abstract

The invention requests to protect a simple easing game solving method based on pruning search, which is characterized by comprising the following steps of: 101, establishing a decision tree, 102, establishing a hash table, 103, carrying out data preprocessing, 104, establishing a strategy diagram, 105, searching the decision tree; 106, searching a cache value; Mathematical principles of simple pier eating games are studied. The method comprises the steps of establishing a pruning tree which can be accurately solved, traversing all low-order instances in a given range through recursion, establishing a Hash cache table, performing data ground preprocessing and data structure equivalent conversion, accurately solving an optimal solution of the given instance, and greatly improving the time efficiency compared with a naive maximum and minimum search. The simple pier-eating game is a true child game of classic pier-eating games such as bridges and Tarot, so that the method can help bridge or Tarot players to more accurately analyze strategies during the game, and can help computer intelligent programs of the bridges or the Tarot reduce the search space, thereby improving the time efficiency of the bridges or the Tarot.

Description

technical field [0001] The invention belongs to the technical fields of game theory, computer game, set theory, and artificial intelligence, and especially constructs accurate decision tree, hash cache, feature construction, model design, process design, etc. through mathematical reasoning. Background technique [0002] Computer games (also called machine games) are one of the most challenging research directions in the field of artificial intelligence. Card game is an intellectual activity that human beings have been engaged in for a long time. The research and cognition of its generics can not only improve human intelligence, but also have high guiding value for the research of industrial production, economic activities and other science and technology. [0003] The development of computer card games is generally inferior to that of chess, because card games basically have incomplete information. Before computers have moderate artificial intelligence, the strategy analysis...

AI Technical Summary

Problems solved by technology

At present, the solving methods of computer double-dummy bridge mainly include hash cache, machine learning, deep learning and their combinations. They have a common defect: if the feature dimension is generalized, only approximate solutions can be obtained in polynomial time.

Whether there is a polynomial algorithm for the generalized solution has not been proved or disproved at present

Images