Intelligent learning in a game-based simulation environment
Saved in:
| Title: | Intelligent learning in a game-based simulation environment |
|---|---|
| Authors: | Gross, David L. |
| Committee Members: | Gonzalez, Avelino J. |
| Summary: | A significant portion of our knowledge of what computers can and cannot do is determined from computer game-playing experiments. In fact one of the earliest performance tests for computers was the game of chess. This was due primarily to the fact that games of strategy provided microworlds of the proper levels of complexity. In addition the progress could be easily measured by playing performance and rank. One major problem in developing computer opponents for games has been the simulating of intelligent playing. The earlier game-playing programs used a non-intelligent bruteforce search method (straight traversal of the game tree). This method was effective for playing finite games of perfect information (i.e., tic-tac-toe), but was ineffective for most other types of strategy games. Later programs used AI techniques in pattern recognition and decision theory and provided only a limited capacity for intelligent playing. The purpose of this thesis is to describe the studies and procedures used to develop a method of simulating intelligent opponent playing for imperfect information games. The domain used for this study will be in the area of physical sports, and in particular in the game of football. The program will have the capacity to learn and adapt to an opponent's strategy and to respond to unanticipated situations (not in its knowledge base) . |
| URL: | https://stars.library.ucf.edu/rtd/3840 |
| Database: | OpenDissertations |
| Abstract: | A significant portion of our knowledge of what computers can and cannot do is determined from computer game-playing experiments. In fact one of the earliest performance tests for computers was the game of chess. This was due primarily to the fact that games of strategy provided microworlds of the proper levels of complexity. In addition the progress could be easily measured by playing performance and rank. One major problem in developing computer opponents for games has been the simulating of intelligent playing. The earlier game-playing programs used a non-intelligent bruteforce search method (straight traversal of the game tree). This method was effective for playing finite games of perfect information (i.e., tic-tac-toe), but was ineffective for most other types of strategy games. Later programs used AI techniques in pattern recognition and decision theory and provided only a limited capacity for intelligent playing. The purpose of this thesis is to describe the studies and procedures used to develop a method of simulating intelligent opponent playing for imperfect information games. The domain used for this study will be in the area of physical sports, and in particular in the game of football. The program will have the capacity to learn and adapt to an opponent's strategy and to respond to unanticipated situations (not in its knowledge base) . |
|---|
