In the field of Artificial Intelligence, there are many studies of reasoning method with uncertainty. Recently, research about the reasoning method from observed fact with uncertainty has been developed, and also the reasoning method which distinguishes the kind of uncertainty is proposed by Matsushima et al. Their research shows clarifying the mathematical model of reasoning method with uncertainty in the statistical standpoint. In this paper, we propose a reasoning system model and a method including uncertainty, based on fuzzy theory. Usually, uncertainty includes not only randomness in field of statistics but also vagueness based on the human's subjective judge with fuzziness. We consider the system in which value of observed fact is calculated using concept of fuzzy events. Moreover, we propose an algorithm of deductive reasoning using Newton-Raphson method, for the reasoning problems with randomness or fuzziness in the system.
In this paper, in order to improve rejection capabilities of the paper currency recognition system for unknown currency patterns on promise of ensuring recognition capabilities for known currency patterns, a feed-forward neural network (FNN) with Gaussian activation function is proposed. The proposed activation function is a ridge-like function. Moreover, a hybrid-learning algorithm for optimizing the width parameters of the Gaussian function is proposed. In the network the Gaussian activation function instead of the sigmoid function is employed in all units of hidden and output layers. The algorithm consists of two steps, one is exploring local minima by employing the gradient descent search, and the other is extricating the search from local minima, in which a random search with the downhill simplex method is employed. The results of simulation reveal the potential effectiveness of the proposed activation function and the algorithm. The system with the proposed activation function and the proposed algorithm can recognize known currency patterns and reject the unknown currency patterns effectively.
A command shaping procedure for sampled-data servo systems is proposed. The command input and the initial state of discrete-time components of the system are determined by solving LMIs to optimize the quadratic tracking performance under the constraints of the values of signals such as input and state saturation. The intersample behavior of the systems is taken into account for both the performance and the constraints.