Takashi Kawakami

A Fundamental Study of a Computer Player Giving Fun To the Opponent: Targeting Hanafuda, a Card Game in Japan

In this research, we aim to create a computer player that gives fun to the opponent. Research on game AI has spread widely in recent years, and many games are being studied. Some of those studies have made remarkable results. Game research is aimed at strengthening computer players. However, it is unknown whether a computer player who is too strong is good. There may also be opponents who think that a computer player is not interesting if it is too strong. Therefore, we thought whether we could create a computer player who entertain the opponent while maintaining a certain degree of strength. To realize this idea, we use the Monte Carlo Tree Search. We tried to create a computer player that gives fun to the opponent by improving the Monte Carlo Tree Search. As a result of some experiments, we succeeded in giving fun, although it was a first step. On the other hand, many problems were found through experiments. In future, it is necessary to solve these problems.

Gradual learning for behavior acquisition by evolving artificial neural network

This paper describes a behavior acquisition of virtual robots by evolving artificial neural network (EANN) with a gradual learning. The gradual learning is a method in which initial states of simulation for evaluation is changing as optimization progresses. Motion of virtual robot is calculated by the physical engine PhysX, and it is controlled by an ANN. Parameters of an ANN are optimized by particle swarm optimization (PSO) so that a virtual robot follows the given target. Experimental results show that the gradual learning is better than a common learning method, realizing the standing behaviors which are not acquired by a common learning at all. It is also shown that random initialization of solutions in the middle of optimization leads to better behaviors.

An Associative Memorization Architecture of Extracted Musical Features from Audio Signals by Deep Learning Architecture

Abstract In this paper, we develop associative memorization architecture of the musical features from time sequential data of the music audio signals. This associative memorization architecture is constructed by using deep learning architecture. Challenging purpose of our research is the development of the new composition system that automatically creates a new music based on some existing music. How does a human composer make musical compositions or pieces? Generally speaking, music piece is generated by the cyclic analysis process and re-synthesis process of musical features in music creation procedures. This process can be simulated by learning models using Artificial Neural Network (ANN) architecture. The first and critical problem is how to describe the music data, because, in those models, description format for this data has a great influence on learning performance and function. Almost of related works adopt symbolic representation methods of music data. However, we believe human composers never treat a music piece as a symbol. Therefore raw music audio signals are input to our system. The constructed associative model memorizes musical features of music audio signals, and regenerates sequential data of that music. Based on experimental results of memorizing music audio data, we verify the performances and effectiveness of our system.

Visualization of Spatial Expression Ability Using a 3-D Expression System

The objective of this paper is to analyze how children develop their spatial expression ability in 3-D space. We prepared a 3-D expression system that has “Translation-test”, “Rotation-test” and “Paper-lay-out-test”. As an experiment, we collected data by using our 3-D expression system for children between the ages of 4 and 6 in a pre-school in August 2009, January 2010, and August 2010. We conducted a survey on correlation between the system and WPPSI: WPPSI is an IQ test. At these experiments, we conducted 5 performance tests with WPPSI. The results are summarized as follows: 1) Children are able to recognize virtual 3-D space of the system, and are able to express images in virtual 3-D space; 2) The data shows the trend that spatial expression ability in space becomes better with advancing ages; 3) We obtained correlations between the system and WPPSI. The system can analyze some abilities that are difficult for WPPSI.