首页 > 最新文献

2019 International Conference on Machine Learning and Cybernetics (ICMLC)最新文献

英文 中文
Development of Rule-Based Agents for Autonomous Parking Systems by Association Rules Mining 基于关联规则挖掘的自主泊车系统规则代理开发
Pub Date : 2019-07-01 DOI: 10.1109/ICMLC48188.2019.8949201
Xin Yuan, M. Liebelt, Peng Shi, B. Phillips
Association Rules Mining is an approach to discover rules from data sets, and it can establish relationships among elements in a data set. Our research is focused on rule-based agents with Artificial General Intelligence (AGI), which are developed based on the overall environment to achieve functions with cognition. In this paper, we use a modified Association Rules Mining method to find out characteristic rules from data recorded in the training of customized parking scenarios. Fuzzy symbolic elements are recorded during training, and Association Rule Mining selects rules for the AI agent. Experiments have been conducted in a virtual environment to demonstrate the effectiveness of the proposed new algorithm.
关联规则挖掘是一种从数据集中发现规则的方法,它可以建立数据集中元素之间的关系。我们的研究重点是基于规则的具有人工通用智能(AGI)的智能体,它是基于整体环境开发的,以实现具有认知的功能。本文采用一种改进的关联规则挖掘方法,从定制停车场景训练中记录的数据中找出特征规则。在训练过程中记录模糊符号元素,关联规则挖掘为人工智能代理选择规则。在虚拟环境中进行了实验,验证了该算法的有效性。
{"title":"Development of Rule-Based Agents for Autonomous Parking Systems by Association Rules Mining","authors":"Xin Yuan, M. Liebelt, Peng Shi, B. Phillips","doi":"10.1109/ICMLC48188.2019.8949201","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949201","url":null,"abstract":"Association Rules Mining is an approach to discover rules from data sets, and it can establish relationships among elements in a data set. Our research is focused on rule-based agents with Artificial General Intelligence (AGI), which are developed based on the overall environment to achieve functions with cognition. In this paper, we use a modified Association Rules Mining method to find out characteristic rules from data recorded in the training of customized parking scenarios. Fuzzy symbolic elements are recorded during training, and Association Rule Mining selects rules for the AI agent. Experiments have been conducted in a virtual environment to demonstrate the effectiveness of the proposed new algorithm.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133643236","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
An Empirical Study on the Classification of Chinese News Articles by Machine Learning and Deep Learning Techniques 基于机器学习和深度学习技术的中文新闻分类实证研究
Pub Date : 2019-07-01 DOI: 10.1109/ICMLC48188.2019.8949309
Chuen-Min Huang, Yi-Jun Jiang
This study compares Chinese news classification results of machine learning (ML) and deep learning (DL). In processing ML, we chose Support Vector Machine (SVM) and Naive Bayes (NB) to form three models: Word2Vec-SVM, TFIDF-SVM, and TFIDF-NB. Since NB assumes that the words are independent, this is different from the concept of related word distribution in Word2Vec, so the combination with NB is excluded. In processing DL, we adopted Bidirectional Long Short-Term Memory (Bi-LSTM), Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN), and used Word2Vec for word embedding. Experimental results showed that with proper word preprocessing, the difference of classification accuracy of ML and DL models is actually very small. Although the results show that Bi-LSTM performs the most accurate and has the lowest Loss compared to other DL techniques, its implementation process is the most time consuming. This study affirms the excellent results of CNN, while its Loss is the highest of the DL models. We also found that Word2Vec-SVM was superior to TFIDF-SVM in terms of efficiency, but its accuracy is not as good as expected. To summarize the classification accuracy in Bi-LSTM, LSTM, CNN, Word2vec-SVM, TFIDF-SVM, and NB are 89.3%, 88%, and 87.54%, 85.32%, 87.35%, 86.56%, respectively.
本研究比较了机器学习(ML)和深度学习(DL)的中文新闻分类结果。在ML处理中,我们选择支持向量机(SVM)和朴素贝叶斯(NB),形成Word2Vec-SVM、TFIDF-SVM和TFIDF-NB三个模型。由于NB假设单词是独立的,这与Word2Vec中相关单词分布的概念不同,因此排除了与NB的结合。在深度学习处理中,我们采用了双向长短期记忆(Bi-LSTM)、长短期记忆(LSTM)和卷积神经网络(CNN),并使用Word2Vec进行词嵌入。实验结果表明,通过适当的词预处理,ML和DL模型的分类准确率差异实际上很小。虽然结果表明,与其他深度学习技术相比,Bi-LSTM的准确率最高,损耗最小,但其实现过程耗时最长。本研究肯定了CNN的优异效果,而其Loss是DL模型中最高的。我们还发现,Word2Vec-SVM在效率上优于TFIDF-SVM,但准确率不如预期。综上所示,Bi-LSTM、LSTM、CNN、Word2vec-SVM、TFIDF-SVM、NB的分类准确率分别为89.3%、88%和87.54%、85.32%、87.35%、86.56%。
{"title":"An Empirical Study on the Classification of Chinese News Articles by Machine Learning and Deep Learning Techniques","authors":"Chuen-Min Huang, Yi-Jun Jiang","doi":"10.1109/ICMLC48188.2019.8949309","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949309","url":null,"abstract":"This study compares Chinese news classification results of machine learning (ML) and deep learning (DL). In processing ML, we chose Support Vector Machine (SVM) and Naive Bayes (NB) to form three models: Word2Vec-SVM, TFIDF-SVM, and TFIDF-NB. Since NB assumes that the words are independent, this is different from the concept of related word distribution in Word2Vec, so the combination with NB is excluded. In processing DL, we adopted Bidirectional Long Short-Term Memory (Bi-LSTM), Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN), and used Word2Vec for word embedding. Experimental results showed that with proper word preprocessing, the difference of classification accuracy of ML and DL models is actually very small. Although the results show that Bi-LSTM performs the most accurate and has the lowest Loss compared to other DL techniques, its implementation process is the most time consuming. This study affirms the excellent results of CNN, while its Loss is the highest of the DL models. We also found that Word2Vec-SVM was superior to TFIDF-SVM in terms of efficiency, but its accuracy is not as good as expected. To summarize the classification accuracy in Bi-LSTM, LSTM, CNN, Word2vec-SVM, TFIDF-SVM, and NB are 89.3%, 88%, and 87.54%, 85.32%, 87.35%, 86.56%, respectively.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114236310","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 3
Using Non-Parametric Regression Methods to Analyze the Impact of air Pollutants on Psychiatric & Neurological Illnesses 使用非参数回归方法分析空气污染物对精神和神经系统疾病的影响
Pub Date : 2019-07-01 DOI: 10.1109/ICMLC48188.2019.8949326
P. Tseng, Fu-Yi Yang, Meng-Han Yang
While industrial pollutions cause changes in the environment and gradually has a strong impact on human physiologies, the relationship between air pollutants and disease occurrences is a subject worthy of exploration. Therefore, based on the nationwide datasets, this study would use non-parametric regression methods to analyze the impact of air pollutants on various psychiatric & neurological illnesses. Through these regression models, the time lag effect of environmental factors on the target diseases would also be taken into account. According to the evaluation outcomes of correlation coefficients, the targets diseases were mainly associated with air pressure, CH4, and SO2. Moreover, observing the coefficients of non-parametric regression models, influences from the environmental factors, i.e. meteorological items and air pollutants, were not limited to the current occurrence (0~1-day lag) but might also accumulate after a period of time (5~7-day lag). In summary, the relationships between air pollutants and psychiatric/neurological illnesses have been verified in this study.
工业污染引起环境变化并逐渐对人类生理产生强烈影响,而空气污染物与疾病发生的关系是一个值得探索的课题。因此,本研究将基于全国数据集,采用非参数回归方法分析空气污染物对各种精神和神经系统疾病的影响。通过这些回归模型,还可以考虑环境因素对目标疾病的时滞效应。从相关系数评价结果来看,目标疾病主要与气压、CH4、SO2相关。此外,观察非参数回归模型的系数,来自环境因素,即气象项目和空气污染物的影响不仅限于当前发生(0~1天滞后),而且可能在一段时间后(5~7天滞后)积累。总之,空气污染物与精神/神经疾病之间的关系已在本研究中得到证实。
{"title":"Using Non-Parametric Regression Methods to Analyze the Impact of air Pollutants on Psychiatric & Neurological Illnesses","authors":"P. Tseng, Fu-Yi Yang, Meng-Han Yang","doi":"10.1109/ICMLC48188.2019.8949326","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949326","url":null,"abstract":"While industrial pollutions cause changes in the environment and gradually has a strong impact on human physiologies, the relationship between air pollutants and disease occurrences is a subject worthy of exploration. Therefore, based on the nationwide datasets, this study would use non-parametric regression methods to analyze the impact of air pollutants on various psychiatric & neurological illnesses. Through these regression models, the time lag effect of environmental factors on the target diseases would also be taken into account. According to the evaluation outcomes of correlation coefficients, the targets diseases were mainly associated with air pressure, CH4, and SO2. Moreover, observing the coefficients of non-parametric regression models, influences from the environmental factors, i.e. meteorological items and air pollutants, were not limited to the current occurrence (0~1-day lag) but might also accumulate after a period of time (5~7-day lag). In summary, the relationships between air pollutants and psychiatric/neurological illnesses have been verified in this study.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116466859","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Class Size Variance Minimization to Metric Learning for Dish Identification 用公制学习最小化班级规模方差来识别菜肴
Pub Date : 2019-07-01 DOI: 10.1109/ICMLC48188.2019.8949253
Shilong Feng, H. Xie, Hongbo Yin, Xiaopeng Chen, Deshun Yang, P. Chan
The objective of metric learning is to search a suitable metric for measuring distance or similarity between samples. Usually, it aims to minimize the distance between samples of same class and maximizes the distance between samples of different classes. However, most metric learning methods do not consider the sizes of classes, which may cause negative impact on the performance in classification since the size of a cluster is usually ignored in the distance comparison. In this work, we propose a triplet loss with variance constraint. Our method focuses not only on the distances between samples but also on the sizes of classes. The size difference between classes is also minimized in our objective function. The experimental results confirm that our method outperforms the one without the class size variance.
度量学习的目的是寻找一个合适的度量来度量样本之间的距离或相似性。通常,它的目标是最小化同类样本之间的距离,最大化不同类样本之间的距离。然而,大多数度量学习方法没有考虑类的大小,这可能会对分类性能产生负面影响,因为在距离比较中通常会忽略聚类的大小。在这项工作中,我们提出了一种具有方差约束的三重态损失。我们的方法不仅关注样本之间的距离,还关注类的大小。在我们的目标函数中,类之间的大小差异也被最小化。实验结果证实了我们的方法优于没有类大小方差的方法。
{"title":"Class Size Variance Minimization to Metric Learning for Dish Identification","authors":"Shilong Feng, H. Xie, Hongbo Yin, Xiaopeng Chen, Deshun Yang, P. Chan","doi":"10.1109/ICMLC48188.2019.8949253","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949253","url":null,"abstract":"The objective of metric learning is to search a suitable metric for measuring distance or similarity between samples. Usually, it aims to minimize the distance between samples of same class and maximizes the distance between samples of different classes. However, most metric learning methods do not consider the sizes of classes, which may cause negative impact on the performance in classification since the size of a cluster is usually ignored in the distance comparison. In this work, we propose a triplet loss with variance constraint. Our method focuses not only on the distances between samples but also on the sizes of classes. The size difference between classes is also minimized in our objective function. The experimental results confirm that our method outperforms the one without the class size variance.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116697360","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Rhyming Knowledge-Aware Deep Neural Network for Chinese Poetry Generation 汉语诗歌生成的韵律知识感知深度神经网络
Pub Date : 2019-07-01 DOI: 10.1109/ICMLC48188.2019.8949208
Wen-Chao Yeh, Yung-Chun Chang, Yu-Hsuan Li, Wei-Chieh Chang
Analyzing and capturing the spirit in the historic Tang Dynasty poems for creating a machine that can compose new poetry is a difficult but fun challenge. In this research, we propose a rhyming knowledge-aware deep neural network for Chinese poetry generation. The model fuses rhyming knowledge that represents phonological tones into a long short-term memory (LSTM) model. This work will help us understand more about what kind of mechanism within the neural network contributes to different styles of the generated poems. The experimental results demonstrate that the proposed method is able to guide the style of those poems towards higher phonological compliance, fluency, coherence, and meaningfulness, as evaluated by human experts. We believe that future research can adopt our approach to further integrate more knowledge such as sentiments, POS, and even stylistic patterns found in poems by famous poets into poem generation.
分析和捕捉唐代古诗中的精神,创造出一种可以创作新诗的机器,这是一项困难但有趣的挑战。在本研究中,我们提出了一个用于汉语诗歌生成的押韵知识感知深度神经网络。该模型将代表语音的押韵知识融合到一个长短期记忆(LSTM)模型中。这项工作将有助于我们更多地了解神经网络内部是什么样的机制促成了不同风格的诗歌生成。实验结果表明,该方法能够引导这些诗歌的风格朝着更高的语音顺应性,流畅性,连贯性和意义的方向发展,正如人类专家所评估的那样。我们相信,未来的研究可以采用我们的方法,将更多著名诗人诗歌中的情感、词性、甚至文体模式等知识进一步整合到诗歌生成中。
{"title":"Rhyming Knowledge-Aware Deep Neural Network for Chinese Poetry Generation","authors":"Wen-Chao Yeh, Yung-Chun Chang, Yu-Hsuan Li, Wei-Chieh Chang","doi":"10.1109/ICMLC48188.2019.8949208","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949208","url":null,"abstract":"Analyzing and capturing the spirit in the historic Tang Dynasty poems for creating a machine that can compose new poetry is a difficult but fun challenge. In this research, we propose a rhyming knowledge-aware deep neural network for Chinese poetry generation. The model fuses rhyming knowledge that represents phonological tones into a long short-term memory (LSTM) model. This work will help us understand more about what kind of mechanism within the neural network contributes to different styles of the generated poems. The experimental results demonstrate that the proposed method is able to guide the style of those poems towards higher phonological compliance, fluency, coherence, and meaningfulness, as evaluated by human experts. We believe that future research can adopt our approach to further integrate more knowledge such as sentiments, POS, and even stylistic patterns found in poems by famous poets into poem generation.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116732441","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 7
Advanced Convolutional Neural Network With Feedforward Inhibition 具有前馈抑制的高级卷积神经网络
Pub Date : 2019-07-01 DOI: 10.1109/ICMLC48188.2019.8949229
Lu Liu, Shuling Yang, D. Shi
Convolutional neural network is a multi-layer neural network with robust pattern recognition ability. However, when the activation function is sigmoid, the convolutional neural network produces gradient vanishing problem. First, this paper analyzes the gradient vanishing problem, and then based on the balance of excitation and inhibition mechanism in neurology, it is proposed to use feed-forward inhibition to reduce activition value and wipe off the scale effect of weights, so that the model can accelerate convergence under the premise of maintaining the nonlinear fitting ability. The results show that the improved convolutional neural network can effectively relieve the gradient vanishing problem.
卷积神经网络是一种具有鲁棒模式识别能力的多层神经网络。然而,当激活函数为s型时,卷积神经网络会产生梯度消失问题。本文首先分析了梯度消失问题,然后基于神经学中激励与抑制机制的平衡,提出利用前馈抑制降低激活值,消除权值的尺度效应,使模型在保持非线性拟合能力的前提下加速收敛。结果表明,改进的卷积神经网络可以有效地缓解梯度消失问题。
{"title":"Advanced Convolutional Neural Network With Feedforward Inhibition","authors":"Lu Liu, Shuling Yang, D. Shi","doi":"10.1109/ICMLC48188.2019.8949229","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949229","url":null,"abstract":"Convolutional neural network is a multi-layer neural network with robust pattern recognition ability. However, when the activation function is sigmoid, the convolutional neural network produces gradient vanishing problem. First, this paper analyzes the gradient vanishing problem, and then based on the balance of excitation and inhibition mechanism in neurology, it is proposed to use feed-forward inhibition to reduce activition value and wipe off the scale effect of weights, so that the model can accelerate convergence under the premise of maintaining the nonlinear fitting ability. The results show that the improved convolutional neural network can effectively relieve the gradient vanishing problem.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114447251","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Using Neural Networks to Label Rain Warning for Natural Hazard of Slope 神经网络在边坡自然灾害降雨预警中的应用
Pub Date : 2019-07-01 DOI: 10.1109/ICMLC48188.2019.8949267
Cheng-Yuan Tang, Whei-Wen Cheng, Tzu-Yen Hsu, C. Jeng, Yi-Leh Wu
The landslides and flows cause significant direct damage to lives and property. A system for monitoring these signs can be the most powerful tool for disaster prevention. In the natural hazard of slope, the signs for rain warning is very useful for disaster prevention. Labeling the rain warning seems to be an important and useful job for disaster prevention. In this paper, two neural network models are used for labeling the rain warning. These two models are the multilayer perceptron (MLP) and the long short-term memory (LSTM). The raw data consist of four observations such as time (time), rainfall (rain), groundwater level (W1) and displacements of inclinometers (SAA-11 and SAA-20). The RMSE (Root Mean Squared Error) using LSTM is 0.161 and RMSE using MLP is 0.212. In the experimental results, LSTM is better than MLP.
山体滑坡和泥石流对生命财产造成重大直接损失。监测这些迹象的系统可能是预防灾害的最有力工具。在边坡自然灾害中,雨水预警标志在防灾中具有重要作用。标注降雨预警似乎是一项重要而有用的防灾工作。本文采用两种神经网络模型对降雨预警进行标注。这两种模型分别是多层感知器(MLP)和长短期记忆(LSTM)。原始数据包括时间(time)、降雨(rain)、地下水位(W1)和倾角仪(SAA-11和SAA-20)位移4个观测值。使用LSTM的RMSE(均方根误差)为0.161,使用MLP的RMSE为0.212。在实验结果中,LSTM优于MLP。
{"title":"Using Neural Networks to Label Rain Warning for Natural Hazard of Slope","authors":"Cheng-Yuan Tang, Whei-Wen Cheng, Tzu-Yen Hsu, C. Jeng, Yi-Leh Wu","doi":"10.1109/ICMLC48188.2019.8949267","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949267","url":null,"abstract":"The landslides and flows cause significant direct damage to lives and property. A system for monitoring these signs can be the most powerful tool for disaster prevention. In the natural hazard of slope, the signs for rain warning is very useful for disaster prevention. Labeling the rain warning seems to be an important and useful job for disaster prevention. In this paper, two neural network models are used for labeling the rain warning. These two models are the multilayer perceptron (MLP) and the long short-term memory (LSTM). The raw data consist of four observations such as time (time), rainfall (rain), groundwater level (W1) and displacements of inclinometers (SAA-11 and SAA-20). The RMSE (Root Mean Squared Error) using LSTM is 0.161 and RMSE using MLP is 0.212. In the experimental results, LSTM is better than MLP.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116567200","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
A Basic Study on Railway Facility Extraction Using a Single-Shot Multi-Box Detector 单发多盒探测器提取铁路设施的基础研究
Pub Date : 2019-07-01 DOI: 10.1109/ICMLC48188.2019.8949296
Masami Nakamura, Yuta Aoto, Shunji Maeda
In railway facilities, there are numerous types and electric train-line facilities. It is difficult to visually inspect all of them, so automatic visual inspection is expected. To achieve automatic inspection, it is important to extract and diagnose the target facilities. This study focuses on facilities extraction by utilizing single-shot multi-box detector (SSD), which can be used as a discriminator for human, car and boat object detection, etc. Diagnosis using Local Subspace Classifier (LSC) is proposed. Herein, we present the evaluation results and the issues applying SSD to the equipment called hangers connecting overhead lines. Some diagnosis results are also explained.
在铁路设施中,有多种类型和电气化的列车线路设施。目视检查所有这些是困难的,因此期望自动目视检查。为了实现自动检测,对目标设施进行提取和诊断是非常重要的。本研究的重点是利用单次多盒探测器(SSD)提取设施,该探测器可作为人、车、船等物体检测的鉴别器。提出了基于局部子空间分类器(LSC)的诊断方法。在此,我们提出了评估结果和将固态硬盘应用于连接架空线路的吊架设备的问题。对一些诊断结果也作了解释。
{"title":"A Basic Study on Railway Facility Extraction Using a Single-Shot Multi-Box Detector","authors":"Masami Nakamura, Yuta Aoto, Shunji Maeda","doi":"10.1109/ICMLC48188.2019.8949296","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949296","url":null,"abstract":"In railway facilities, there are numerous types and electric train-line facilities. It is difficult to visually inspect all of them, so automatic visual inspection is expected. To achieve automatic inspection, it is important to extract and diagnose the target facilities. This study focuses on facilities extraction by utilizing single-shot multi-box detector (SSD), which can be used as a discriminator for human, car and boat object detection, etc. Diagnosis using Local Subspace Classifier (LSC) is proposed. Herein, we present the evaluation results and the issues applying SSD to the equipment called hangers connecting overhead lines. Some diagnosis results are also explained.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122137518","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Mobility-Based Clustering With Link Quality Estimation for Urban Vanets 基于交通出行的城市道路质量聚类方法
Pub Date : 2019-07-01 DOI: 10.1109/ICMLC48188.2019.8949241
H. Ferng, Muhammad Abdullah
Owing to high mobility and a large amount of vehicles in a vehicular ad hoc network (VANET), it is challenging to overcome the issues of frequent topology changes and network scalability. To mitigate these issues, a vehicle clustering and management scheme can be applied to VANETs. Towards this goal, a mobility-based clustering scheme with a clustering link quality estimation (CLQE) metric considering both mobility information and link quality estimation is proposed in this paper. The proposed clustering scheme is evaluated through the NS-3 simulator and our simulation results show that our proposed scheme outperforms the closely related schemes in most scenarios.
由于车辆自组织网络(VANET)具有高移动性和大量的车辆,因此克服频繁的拓扑变化和网络可扩展性问题具有挑战性。为了缓解这些问题,车辆集群和管理方案可以应用于VANETs。为此,提出了一种同时考虑移动性信息和链路质量估计的基于移动性的聚类方案。通过NS-3模拟器对所提出的聚类方案进行了评估,仿真结果表明,所提出的聚类方案在大多数情况下优于密切相关的聚类方案。
{"title":"Mobility-Based Clustering With Link Quality Estimation for Urban Vanets","authors":"H. Ferng, Muhammad Abdullah","doi":"10.1109/ICMLC48188.2019.8949241","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949241","url":null,"abstract":"Owing to high mobility and a large amount of vehicles in a vehicular ad hoc network (VANET), it is challenging to overcome the issues of frequent topology changes and network scalability. To mitigate these issues, a vehicle clustering and management scheme can be applied to VANETs. Towards this goal, a mobility-based clustering scheme with a clustering link quality estimation (CLQE) metric considering both mobility information and link quality estimation is proposed in this paper. The proposed clustering scheme is evaluated through the NS-3 simulator and our simulation results show that our proposed scheme outperforms the closely related schemes in most scenarios.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122284343","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 3
Behavior Acquisition on a Mobile Robot Using Reinforcement Learning With Continuous State Space 基于连续状态空间强化学习的移动机器人行为获取
Pub Date : 2019-07-01 DOI: 10.1109/ICMLC48188.2019.8949181
T. Arai, Y. Toda, N. Kubota
In the application of Reinforcement Learning to real tasks, the construction of state space is a significant problem. In order to use in the real-world environment, we need to deal with the problem of continuous information. Therefore, we proposed a method of the construction of state space using Growing Neural Gas. In our method, the agent constructs a state space model from its own experience autonomously. Furthermore, it can reconstruct the suitable state space model to adapt the complication of the environment. Through the experiments, we showed that Reinforcement Learning could be performed efficiently by successively updating the state space model according to the environment.
在将强化学习应用于实际任务中,状态空间的构建是一个重要的问题。为了在现实环境中使用,我们需要处理连续信息的问题。因此,我们提出了一种利用生长神经气体构造状态空间的方法。在我们的方法中,智能体根据自己的经验自主构建状态空间模型。此外,该方法还能重构出适合的状态空间模型,以适应环境的复杂性。通过实验,我们证明了根据环境逐次更新状态空间模型可以有效地进行强化学习。
{"title":"Behavior Acquisition on a Mobile Robot Using Reinforcement Learning With Continuous State Space","authors":"T. Arai, Y. Toda, N. Kubota","doi":"10.1109/ICMLC48188.2019.8949181","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949181","url":null,"abstract":"In the application of Reinforcement Learning to real tasks, the construction of state space is a significant problem. In order to use in the real-world environment, we need to deal with the problem of continuous information. Therefore, we proposed a method of the construction of state space using Growing Neural Gas. In our method, the agent constructs a state space model from its own experience autonomously. Furthermore, it can reconstruct the suitable state space model to adapt the complication of the environment. Through the experiments, we showed that Reinforcement Learning could be performed efficiently by successively updating the state space model according to the environment.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126012690","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
期刊
2019 International Conference on Machine Learning and Cybernetics (ICMLC)
全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1