Showing papers by "Nobuyuki Matsui published in 2005"

Qubit neural network and its learning efficiency

Abstract: Neural networks have attracted much interest in the last two decades for their potential to realistically describe brain functions, but so far they have failed to provide models that can be simulated in a reasonable time on computers; rather they have been limited to toy models. Quantum computing is a possible candidate for improving the computational efficiency of neural networks. In this framework of quantum computing, the Qubit neuron model, proposed by Matsui and Nishimura, has shown a high efficiency in solving problems such as data compression. Simulations have shown that the Qubit model solves learning problems with significantly improved efficiency as compared to the classical model. In this paper, we confirm our previous results in further detail and investigate what contributes to the efficiency of our model through 4-bit and 6-bit parity check problems, which are known as basic benchmark tests. Our simulations suggest that the improved performance is due to the use of superposition of neural states and the use of probability interpretation in the observation of the output states of the model.

An Examination of Qubit Neural Network in Controlling an Inverted Pendulum

Abstract: The Qubit neuron model is a new non-standard computing scheme that has been found by simulations to have efficient processing abilities. In this paper we investigate the usefulness of the model for a non linear kinetic control application of an inverted pendulum on a cart. Simulations show that a neural network based on Qubit neurons would swing up and stabilize the pendulum, yet it also requires a shorter range over which the cart moves as compared to a conventional neural network model.

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

Abstract: 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.

Self-Organizing Map Based on Block Learning

Abstract: 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.

Reinforcement learning by chaotic exploration generator in target capturing task

Abstract: The exploration, that is a process of trial and error, plays a very important role in reinforcement learning. As a generator for exploration, it seems to be familiar to use the uniform pseudorandom number generator. However, it is known that chaotic source also provides a random-like sequence as like as stochastic source. Applying this random-like feature of deterministic chaos for a generator of the exploration, we already found that the deterministic chaotic generator for the exploration based on the logistic map gives better performances than the stochastic random exploration generator in a nonstationary shortcut maze problem. In this research, in order to make certain such a difference of the performance, we examine target capturing as another nonstationary task. The simulation result in this task approves the result in our previous work.

Perceptual binding by coupled oscillatory neural network

Abstract: The binding problem is a problem on the integration of perceptual properties in our brains. For describing this problem in the artificial neural network, it is necessary to introduce the temporal coding of information. In this paper, we propose a neural network model that can represent the bindings of external stimuli, based on the network that is capable of figure-ground segmentation proposed by Sompolinsky and Tsodyks. This model adopts the coupled oscillators that can represent the temporal coding and the synchronization among them.

On defect-tolerance in cellular computers

Abstract: Defects will be a major problem for manufacturing computers by nanotechnology. This paper explores this issue for nanocomputers based on cellular arrays. Known for their regular structure, such architectures promise cost-effective manufacturing based on molecular self-organization. We show a self-contained way to detect defects in a cellular array, and to configure circuits on its cells while avoiding the defects. Timing is assumed asynchronous, i.e., updating of cells is not controlled by a clock.

On Self-Organizing Map Approaches for Data Detection of Hematopoietic Tumors

Abstract: Data detection using self organizing maps is presented for hematopoietic tumor patients. The learning data for the maps is generated from the screening data. Redundant items, which have an unfavorable influence on data detection and are common to all the data, are eliminated by a genetic algorithm and an immune algorithm. 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 methods achieve the high probability of correctly identifying examinees as hematopoietic tumor patients