Changes

'''TAAI 2018''',<br/>rhe the 2018 Conference on Technologies and Applications of Artificial Intelligence, [https://en.wikipedia.org/wiki/Taichung Taichung], [https://en.wikipedia.org/wiki/Taiwan Taiwan], November 30 - December 2, 2018.

: For the game of [[Go]], [[Chess]], and [[Shogi]] (Japanese Chess), [[Deep Learning|deep neural networks ]] (DNNs) have contributed to build accurate evaluation functions and many studies have attempted to create so called the value network which predicts a reward of a given state. A recent study of the value network for the game of Go has shown that a two-headed neural network with two different objectives can be trained effectively and performs better than a single-headed network. One of the two heads is called a value head and the other, policy head, predicts next moves at a given state. This multitask training makes the network more robust and improves the generalization performance. In this paper we show that a simple discriminator network is an alternative target of the multitask learning. Compared to the existing deep neural network, our proposed network can be designed more easily because of its simple output. Experimental results showed that our discriminative target also makes the learning stable and the evaluation function trained by our method is comparable to the training of existing studies in terms of predicting next moves and playing strength.

: [[Monte-Carlo Tree Search|Monte-Carlo tree search]] (MCTS) algorithms play an important role in developing computer players for many games. The performance of MCTS players is often leveraged in combination with offline knowledge, i.e., evaluation functions. In particular, recently [[AlphaGo ]] and [[AlphaGo Zero ]] achieved a big success in developing strong computer [[Go ]] player by combining evaluation functions consisting of [[Deep Learning|deep neural networks ]] with a variant of [[Christopher D. Rosin#PUCT|PUCT ]] (Predictor + [[UCT|UCB applied to trees]]). The effect of evaluation functions on the strength of MCTS algorithms, however, has not been investigated well, especially in terms of the quality of evaluation functions. In this study, we address this issue and empirically analyze the AlphaGo's PUCT algorithm by using [[Othello ]] (Reversi) as the target game. We investigate the strength of PUCT players using variants of an existing evaluation function of a champion-level computer player. From intensive experiments, we found that the PUCT algorithm works very well especially with a good evaluation function and that the value function has more importance than the policy function in the PUCT algorithm.