Skip To Main Content
Image of friends playing a card game
Dr. Xia "Ben" Hu and his team are collaborating with a leading gaming company in China to develop a gaming bot for Sheng ji, one of the most popular Asian card games. | Image: Getty Images

Dr. Xia “Ben” Hu and his graduate students are collaborating with JJ World, a leading gaming company in China, to develop a Sheng ji gaming bot (Sheng ji is one of the most popular Asian card games). The new bot will not only provide a better gaming experience for users by matching their skill set, but will also advance the field of machine learning in environments with incomplete information.

Previous research has already been done on games such as Go (an abstract strategy board game). These games have a set number of moves and player strategies, enabling a computer to compute and search for the next steps. The available information is complete.

“Sheng ji is a typical game with incomplete information,” said Hu, a computer science and engineering assistant professor at Texas A&M University. “When a player makes a decision, she needs to consider the hidden cards from the other three players. To tackle the information asymmetry we need to implement a model to predict the hidden cards from the other players.”

Hu said the system is expected to be divided into three modules: the memory module, the prediction module and the decision module. The memory module, as the game progresses, will record all the known information that serves as the input of hands prediction in the prediction module. The outputs of the memory module and prediction will be further used for decision making in the decision module.

Image of Ben Hu
Dr. Xia “Ben” Hu, principal investigator for the project. | Image: Texas A&M University

Hu’s work on Sheng ji represents a high-risk, high-reward opportunity. Hu’s team has the unique opportunity to integrate their research into a real-world gaming system with access to millions of users. Through online tests, they will be able to analyze player behavior and iteratively improve the gaming strategies. The more adaptive the game, the better the player experience will be, enticing players to play more Sheng ji. Their work will also be applicable to other related games.

“Hopefully by collaborating with one of the largest gaming platforms in China, we can develop learning algorithms with impact in this domain,” said Hu.

Hu, along with his graduate students Daochen Zha and Kwei-herng Lai, started this research one year ago in the Data Analytics Texas A&M Laboratory. In December, JJ World gave Hu an industrial grant of $108,000 to advance the algorithms and systems involved.

“Many challenges and opportunities await,” said Hu.