Naotake Kamiura

Defect-tolerance in cellular nanocomputers

For the manufacturing of computers built by nanotechnology, defects are expected to be a major problem. This paper explores this issue for nanocomputers based on cellular automata. Known for their regular structure, such architectures promise cost-effective manufacturing based on molecular self-organization. We show how a cellular automaton can detect defects in a self-contained way, and how it configures circuits on its cells while avoiding the defects. The employed cellular automaton is asynchronous, i.e., it does not require a central clock to synchronize the updates of its cells. This mode of timing is especially suitable for the high integration densities of nanotechnology implementations, since it potentially causes less heat dissipation.

On detection of hematopoietic tumors using self organizing maps and genetic algorithms

This paper proposes the scheme of detecting the screening data of hematopoietic tumor patients, using self-organizing maps. The data of an examinee frequently lacks several of the item values. In addition, there exist redundant common items that should be eliminated from all of the data because they have an unfavorable influence on classifying the data. The data imputation, which substitutes the averages of non-missing item values, and a genetic algorithm are adopted to overcome the above issues. It is basically judged, by observing a label of a winner neuron in a map, whether the data presented to the map belongs to the class of hematopoietic tumors. Quantitative evaluations show that the proposed scheme achieves the high probability of correctly identifying examinees as hematopoietic tumor patients.

Dynamics of discrete-time quaternionic hopfield neural networks

We analyze a discrete-time quaternionic Hopfield neural network with continuous state variables updated asynchronously. The state of a neuron takes quaternionic value which is four-dimensional hypercomplex number. Two types of the activation function for updating neuron states are introduced and examined. The stable states of the networks are demonstrated through an example of small network.

Computing by Swarm Networks

Though the regular and fixed structure of cellular automata greatly contributes to their simplicity, it imposes a strict limitation on the applications that can be modeled by them. This paper proposes swarm networks, a model in which cells, unlike in cellular automata, have irregular neighborhoods. Timed asynchronously, a cell in this model acts like an agentthat can dynamically interact with a varying set of other cells under the control of transition rules. The configurations in which cells are organized according to their neighborhoods can move around in space, following simple mechanical laws. We prove computational universality of this model by simulating a circuit consisting of asynchronously timed circuit modules. The proposed model may find applications in nanorobotic systems and artifical biological systems.

Self-Organizing Map Based on Block Learning

A block-matching-based self-organizing map (BMSOM) is presented. Finding a winner is carried out for each block, which is a set of neurons arranged in square. The proposed learning process updates the reference vectors of all of the neurons in a winner block. Then, the degrees of vector modifications are mainly controlled by the size (i.e., the number of neurons) of the winner block. To prevent a single cluster with neurons from splitting into some disjointed clusters, the restriction of the block size is imposed in the beginning of learning. At the main stage, this restriction is canceled. In BMSOM learning, the size of a winner block does not always decrease monotonically. The formula used to update the reference vectors is basically uncontrolled by time. Therefore, even if a map is in a nonstationary environment, training the map is probably pursued without interruption to adjust time-controlled parameters such as learning rate. Numerical results demonstrate that the BMSOM makes it possible to improve the plasticity of maps in a nonstationary environment and incremental learning.

Online marking of defective cells by random flies

Defect-tolerance, the ability to overcome unreliability of components in a system, will be essential to realize computers built by nanotechnology This paper presents a novel approach to defect-tolerance for nanocomputers that are based on self-timed cellular automata, a type of asynchronous cellular automaton According to this approach, defective cells are detected and isolated by configurations of random flies that move around in cellular space We show that detection and isolation are realized in an on-line manner, i.e., while computation takes place.

Learning based on fault injection and weight restriction for fault-tolerant Hopfield neural networks

Hopfield neural networks tolerating weight faults are presented. The weight restriction and fault injection are adopted as fault-tolerant approaches. For the weight restriction, a range to which values of weights should belong is determined during the learning, and any weight being outside this range is forced to be either its upper limit or lower limit. A status of a fault occurring is then evoked by the fault injection, and calculating weights is made under this status. The learning based on both of the above approaches surpasses the learning based on either of them in the fault tolerance and/or in the learning time.

A neural network approach for counting pedestrians from video sequence images

A system for counting pedestrians in sequence images obtained from single video camera is proposed in this paper. This system has the capabilities of simultaneously detecting and tracking several groups of pedestrians. Groups can be extracted by using the background subtraction method, and a layered neural network with BP learning algorithm is applied to estimate the number of pedestrians in each of the groups. The practical applicability of the proposed system is demonstrated, applying it to the sequence images of a real scenery.

On Self-Organizing Map Based Classification of Insect Neurons

In this paper, a systematic method based on self-organizing maps is presented to classify interneurons of silkworm moths. Denseness of branching structures and existence of thick main dendrites are quantified by six fractal dimension values and three values calculated from images to which fundamental processing techniques are applied, respectively. Such values are employed as nine elements in training data for a map. The classification result is obtained as clusters with units in the trained map. Experimental results establish that the classification executed by the proposed method is comparable in accuracy to the manually executed classification

PODEM based on static testability measures and dynamic testability measures for multiple-valued logic circuits

In this paper, PODEM for multiple-valued logic circuits is proposed. It consists of the D-propagation, implication and backtracing operation. To guide the D-propagation (or backtracing operation), observability (or controllability) measures are introduced. They are computed by simple recursive formulas, and enable us to reduce the frequency of backtracking. In addition, the scheme of exploiting static testability measures up to a certain stage of test generation and then resorting to dynamic testability measures is incorporated into PODEM. The experimental results on ternary benchmark circuits show that the above scheme is useful in generating as many test patterns as possible whilst shortening the total time required for the test generation.