Home | Mysite

Ruiyang Xu

I'm a Ph.D. student at Khoury College of Computer Sciences. I'm doing studies in AI and Reinforcement Learning. To be specific, my research interesting is to solve combinatorial problems through self-play AI algorithms. I'm working on building a general end-to-end game player on such tasks.

Copy%2520of%2520HSR_REC%2520(2)_edited_e

Solving the Highest Safe Rung (HSR) Problem through a neural MCTS self-play Algorithm

In this research, we apply the neural MCTS algorithm (AlphaZero) to solve a prototype combinatorial optimization problem (HSR).

An attempt to solving QSAT problem through Graph Neural Network based neural MCTS

In this study, we try to tackle the QSAT problem using the neural MCTS technique. We propose a novel graph neural network to encode QBFs.

Screen%20Shot%202020-02-21%20at%2012.19_

Building a general game player for problem solving via gamification and neural MCTS

This is an ongoing project where we are going to build a general game player on problem solving. We use predicate logic as game description.

Ruiyang Xu, Ph.D. Student

Ruiyang Xu

RECENT RESEARCHES

Solving the Highest Safe Rung (HSR) Problem through a neural MCTS self-play Algorithm

An attempt to solving QSAT problem through Graph Neural Network based neural MCTS

Building a general game player for problem solving via gamification and neural MCTS