Monte-Carlo Simulation Adjusting

In this paper, we propose a new learning method simulation adjusting that adjusts simulation policy to improve the move decisions of the Monte Carlo method. We demonstrated simulation adjusting for 4 × 4 board Go problems. We observed that the rate of correct answers moderately increased.

Authors

• Nobuo Araki The University of Electro-Communications
• Masakazu Muramatsu The University of Electro-Communications
• Hoki Kunihito The University of Electro-Communications
• Satoshi Takahashi The University of Electro-Communications

Keywords

No keywords are indexed for this paper.