Research on Opponent Modeling in Imperfect Information Games
Journal Title: 河南科技大学学报（自然科学版） - Year 2019, Vol 40, Issue 1
Abstract
For the problem that the traditional explicit modeling relied on large numbers of data samples, the policy bootstrapping algorithm was introduced to improve the modeling efficiency through the bootstrapping of sample data. Meanwhile, in order to enhance the accuracy of opponent model, the implicit modeling method and subpolicy implicit modeling method were combined to propose subpolicy discovery algorithm. The game of Leduc poker was used as an experimental subject to compare two traditional methods and two new algorithms.The results indicate that policy bootstrapping improves the efficiency of explicit modeling and the accuracy of the model. Compared with the explicit modeling method, policy bootstrapping algorithm improves 84. 4% in profits by using the opponent's weakness. The subpolicy discovery algorithm improves 128. 6% compared with the implicit modeling method.
Authors and Affiliations
Tiandong WU, Ying SHI
Keywords
Related Articles
- EP ID EP427032
- DOI 10.15926/j.cnki.issn1672-6871.2019.01.010
- Views 109
- Downloads 0
Microstructure Characteristic and Defects of Alumina Ceramic Formed by Laser Additive Manufacturing
The laser additive manufacturing forming process of alumina ceramic materials was investigated. The internal microstructure characteristics and defects of the formed structure were observed by optical metallographic micr...
Combustion Optimization of a High Speed High Power Marine Diesel Engine
For the problem of excessive smoke and high fuel consumption in the marine high-speed high-power diesel engine, the optimization was conducted based on bench test and combustion simulation. Fuel injection system bench te...
SOC Estimation of Vanadium Redox Flow Battery Based on Improved PNGV Model
In view of the problem that the branch current generated by the circulation pump had an impact on the estimation of state of charge ( SOC) during charging and discharging of vanadium redox flow battery ( VRB) , the estim...
Preprocessing Methods for Mobile Measurements of Roadside Air Pollution
When roadside air pollution was measured by a mobile platform, the measurement was prone to be interfered by abnormal samples of high values, pollution background and spatiotemporal scale. Therefore, data preprocessing m...
Control Strategy of Miniature Pure Electric Truck Based on Driver's Intention Recognition
In order to improve the energy efficiency of miniature pure electric truck, a vehicle control strategy based on driver's intention recognition was proposed.Based on Stateflow finite state machine, a primary driver recogn...