Complete and Compressive Description of Spoken Words and Music Data with use of Cellular Automata and Analysis of Their Complexity

使用元胞自动机对口语和音乐数据进行完整和压缩的描述及其复杂性分析

• 批准号: 13831008
• 负责人: NARA Shigetoshi
• 金额: $ 1.92万
• 依托单位: Okayama University
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (C)
• 财政年份: 2001
• 资助国家: 日本
• 起止时间: 2001 至 2003
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-13831008/
• 关键词: Cellular Automata; Rule Dynamics; Bit Patterns; Information Compression; Spoken Words Music Motion Pictures; Complexity; Compressibility; Signal to Noise Ratio

Concerning the proposed subjects, computer simulations were done and the obtained results were presented in research meetings, Domestic conferences, and International conferences. Also their results were published as scientific papers in several journals. The summary of the results are as follows. Nonlinear dynamics in a neural network model are investigated by means of numerical simulations and a question whether the same dynamical properties is able to be described by rules of cellular automata or not., are discussed. The answer is affirmative and an example is actually shown in a specified case of 2-dimensional motion pictures provided one employs not 2-states-3-neighbors but many neighbors with totalistic rules. Numerical simulations were done and the successful results were reported in domestic conferences and in international conferences as well. Several papers were written by the investigators and a few were already published in a few journals. A third meeting was held with all the investigators together and the recent contents of each research activities were introduced and discussed.

针对所提出的课题，进行了计算机模拟，并将所得结果在研究会议、国内会议和国际会议上展示。他们的研究结果还以科学论文的形式发表在多家期刊上。结果总结如下。通过数值模拟研究神经网络模型中的非线性动力学，并讨论相同的动力学特性是否能够通过元胞自动机规则描述的问题。答案是肯定的，并且实际上在二维电影的特定情况下显示了一个示例，前提是不使用 2 状态 3 邻居，而是使用具有极权规则的许多邻居。进行了数值模拟，并在国内会议和国际会议上报告了成功的结果。研究人员撰写了几篇论文，其中一些已经在一些期刊上发表。全体研究者召开了第三次会议，介绍并讨论了近期各项研究活动的内容。

项目成果

On Some Necessary and Sufficient Conditions for a Recurrent Neural Network Model With Time Delays to Generate Oscillations

时滞循环神经网络模型产生振荡的若干充要条件

• DOI: 10.1109/tnn.2010.2047512
• 发表时间: 2010-08-01
• 期刊: IEEE Transactions on Neural Networks
• 作者: C. Feng;R. Plamondon;C. O’Reilly
• 通讯作者: C. O’Reilly

Can potentially useful dynamics to solve complex problems emerge from constrained chaos and/or chaotic itinerancy?

受限混沌和/或混沌行程能否产生解决复杂问题的潜在有用动力？

• DOI: 10.1063/1.1604251
• 发表时间: 2003-08-22
• 期刊: Chaos
• 影响因子: 2.9
• 作者: S. Nara

A Note on Time Delayed Effect in a Recurrent Neural Network Model

关于循环神经网络模型中时滞效应的说明

• DOI: 10.1007/s00521-003-0349-0
• 发表时间: 2003-06-01
• 期刊: Neural Computing & Applications
• 影响因子: 6
• 作者: Y. Suemitsu;S. Nara
• 通讯作者: S. Nara