A State Unification in Learning Finite State Machines using Recurrent Neural Networks

机译：使用递归神经网络学习有限状态机的状态统一

获取原文

页面导航

摘要
著录项
相似文献
相关主题

摘要

Recurrent Neural Networks (RNNs) have an ability of learning Finite State Machines (FSMs), but the optimal structurizing of the networks is difficult. In the learning of the target FSM, it is important not to generate excessive states because of the possibility of obtaining the minimal FSM. In this paper, we build a method of structurization for the networks with using the genetic algorithm, and using our method we realize a state unification in the state space that means the exclusion the excessive states to get the optimal cycles. Also we derive the mutual information for measuring the degree of the state unification. We show that our method proceeds the state unification in the state space.

机译：经常性神经网络（RNN）具有学习有限状态机（FSM）的能力，但网络的最佳结构难以实现。在目标FSM的学习中，由于获得最小FSM的可能性，重要的是不产生过多的状态。在本文中，我们使用遗传算法构建网络结构化方法，并使用我们的方法，我们在状态空间中实现了一个状态统一，这意味着排除过多状态以获得最佳周期。我们还派生了衡量国家统一程度的相互信息。我们表明我们的方法在国家空间中进入了国家统一。

著录项

来源
《World multiconference on systemics, cybernetics and informatics;SCI 2000》|2000年|p.509-513|共5页
会议地点
作者
Hiroyuki SHIOYA; Keisuke OZAKI; Tsutomu DA-TE;
展开▼
作者单位

展开▼
会议组织
原文格式 PDF
正文语种
中图分类无线电电子学、电信技术;
关键词
recurrent neural network; finite state machine; genetic algorithm; state space of an RNN; mutual information;

机译：递归神经网络有限状态机遗传算法RNN的状态空间;共同信息;

相似文献

外文文献
中文文献
专利

1. Advances in artificial neural networks, machine learning and computational intelligence Selected papers from the 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018) [J] . Oneto Luca, Bunte Kerstin, Schleif Frank-Michael Neurocomputing . 2019,第MAYa21期

机译：人工神经网络，机器学习和计算智能的进步选自第26届欧洲人工神经网络，计算智能和机器学习研讨会（ESANN 2018）的论文
2. Advances in artificial neural networks, machine learning and computational intelligence Selected papers from the 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018) [J] . Oneto Luca, Bunte Kerstin, Schleif Frank-Michael Neurocomputing . 2019,第May21期

机译：人工神经网络，机器学习和计算智能的进步从第26届欧洲人工神经网络，计算智能和机器学习中选择了论文（拍摄2018）
3. Unification neural networks: unification by error-correction learning [J] . Ekaterina Komendantskaya Logic Journal of IGPL . 2011,第6期

机译：统一神经网络：通过纠错学习实现统一
4. On the Interpretation of Recurrent Neural Networks as Finite State Machines [C] . Christian Oliva, Luis F. Lago-Fernandez International Conference on Artificial Neural Networks . 2019

机译：递归神经网络作为有限状态机的解释
5. Recurrent neural network learning and neural network learning controller. [D] . Yan, Lilai. 1994

机译：递归神经网络学习和神经网络学习控制器。
6. EEG-Based Emotion Classification for Alzheimer’s Disease Patients Using Conventional Machine Learning and Recurrent Neural Network Models [O] . Jungryul Seo, Teemu H. Laine, Gyuhwan Oh, 2020

机译：基于EEG的Alzheimer疾病患者的情感分类使用传统机器学习和经常性神经网络模型
7. Learning a Class of Large Finite State Machines with a Recurrent Neural Network [O] . C. Lee Giles, B. G. Horne, T. Lin 1995

机译：用递归神经网络学习一类大型有限状态机

A State Unification in Learning Finite State Machines using Recurrent Neural Networks

摘要

著录项

相似文献

相关主题

期刊订阅