We have proposed a two-level architecture for brain computing, where two levels are introduced for processing of meta-symbol. At Level 1 a conventional pattern recognition is performed, where neural computation is included, and its output gives the meta-symbol which is a symbol enlarged from a symbol to a kind of pattern. At Level 2 an algorithm acquisition is made by using a state machine for abstract states (which is a meta-symbol expression). We are also developing the VLSI chips at each level for the structured brain computer (SBC) version 1.0. To explain SEC we introduce the brain computing and the artificial memory system as well as the hardware of SBC.

The paper investigates the integration of the neural network technique and traditional AI techniques towards the realization of a real-time neuron-based AI architecture. As the first step of our project, we propose a neural network based AI system, called NAI. NAI is a kind of real-time CBR (case-based reasoning) system in which the WTA (winner-take-all) type neural network is embedded for supporting the real-time classification and retrieval of a massive case-base. For flexible learning with the WTA neural network, two learning algorithms (supervised and unsupervised) have been developed on the basis of the LVQ1 and self-organizing learning algorithms.

The distributed k-mutual-exclusion problem (k-mutex problem) is the problem of guaranteeing that at most k processes at a time can enter a critical section at a time in a distribution system. A method proposed for the solution of the distributed mutual exclusion problem (i.e., 1-mutex problem) by D. Barbara and H. Garcia-Molina (1987) is an extension of majority consensus and uses coteries. The goodness of coterie-based 1-mutex algorithm strongly depends on the availability of coterie, and it has been shown that majority coterie is optimal in this sense, provided that: the network topology is a complete graph, the links never fail, and p, the reliability of the process, is at least 1/2. The concept of a k-coterie, an extension of a coterie, is introduced for solving the k-mutex problem, and lower and upper bounds are derived on the reliability p for k-majority coterie, a natural extension of majority coterie, to be optimal, under conditions (1)-(3). For example, when k=3, p must be greater than 0.994 for k-majority coterie to be optimal.>

Three-dimensional VLSI (in short, 3-D VLSI) is a new device technology that is expected to realize high performance systems. In this paper, we propose an image processing architecture based on 3-D VLSI consisting of optically-connected layers. Since the optical inter-layer connection seems to have some of useful functions due to isotropic radiation of the light, we algebraicly formulate them as the picture processing operators. Moreover, we show that the operators are available for applications such as template matching. The availability of a proposed template matching algorithm is verified by simulation.

In this paper we propose a method to implement the moving patterns which are consecutively related in a small computer system. Suppose that the patterns are constructed of many block elements. A pattern is derived by the operation ( such as AND, OR, EXOR, MOVE, ZOOM, ROTATE,. .. ), with another pattern if it is required as operand. The parallel execution on a multiple microprocessor system is also discussed.