ThesisAuthors: Lê, Kim Thư; Advisor: Bùi, Thế Duy (2010)
Luận văn ThS. Công nghệ thông tin -- Trường Đại học Công nghệ. Đại học Quốc gia Hà Nội, 2010; Electronic Resources; Autonomous learning is one of the main reasons of the human development. Many computer systems is also need this properties to improve its performance. Nevertheless, it is very difficult to construct such system. The reinforcement learning framework is the prominent solution for solving the problem. In the thesis, we construct a self-learning system that uses a reinforcement learning algorithm - Q-learning. In addition, we also propose a new method for control choosing action based on supervised learning. Our experiments show that except some bad cases have happened, thi... 
