Pub Date : 2020-05-19DOI: 10.1109/IJCNN.2019.8852367
Shijun Wang, Baocheng Zhu, Lintao Ma, Yuan Qi
In this paper, we consider optimizing a smooth, convex, lower semicontinuous function in Riemannian space with constraints. To solve the problem, we first convert it to a dual problem and then propose a general primal-dual algorithm to optimize the primal and dual variables iteratively. In each optimization iteration, we employ a proximal operator to search optimal solution in the primal space. We prove convergence of the proposed algorithm and show its non-asymptotic convergence rate. By utilizing the proposed primal-dual optimization technique, we propose a novel metric learning algorithm which learns an optimal feature transformation matrix in the Riemannian space of positive definite matrices. Preliminary experimental results on an optimal fund selection problem in fund of funds (FOF) management for quantitative investment showed its efficacy.
{"title":"A Riemannian Primal-dual Algorithm Based on Proximal Operator and its Application in Metric Learning","authors":"Shijun Wang, Baocheng Zhu, Lintao Ma, Yuan Qi","doi":"10.1109/IJCNN.2019.8852367","DOIUrl":"https://doi.org/10.1109/IJCNN.2019.8852367","url":null,"abstract":"In this paper, we consider optimizing a smooth, convex, lower semicontinuous function in Riemannian space with constraints. To solve the problem, we first convert it to a dual problem and then propose a general primal-dual algorithm to optimize the primal and dual variables iteratively. In each optimization iteration, we employ a proximal operator to search optimal solution in the primal space. We prove convergence of the proposed algorithm and show its non-asymptotic convergence rate. By utilizing the proposed primal-dual optimization technique, we propose a novel metric learning algorithm which learns an optimal feature transformation matrix in the Riemannian space of positive definite matrices. Preliminary experimental results on an optimal fund selection problem in fund of funds (FOF) management for quantitative investment showed its efficacy.","PeriodicalId":222797,"journal":{"name":"2019 International Joint Conference on Neural Networks (IJCNN)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133414374","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}
Pub Date : 2019-09-30DOI: 10.1109/IJCNN.2019.8852007
E. Carvalho, Bruno V. Ferreira, G. P. R. Filho, P. Gomes, G. Freitas, P. A. Vargas, J. Ueyama, G. Pessin
This paper proposes a fault-tolerant indoor localization system that employs Recurrent Neural Networks (RNNs) for the localization task. A decision module is designed to detect failures and this is responsible for the allocation of RNNs that are suitable for each situation. As well as the fault-tolerant system, several architectures and models for RNNs are exploited in the system: Gated Recurrent Unit (GRU), Long Short-Term Memory (LSTM) and Simple RNN. The system uses as inputs a collection of Wi-Fi Received Signal Strength Indication (RSSI) signals, and the RNN classifies the position of an agent on the basis of this collection. A fault-tolerant mechanism has been designed to handle two types of failures: (i) momentary failure, and (ii) permanent failure. The results show that the RNNs are suitable for tackling the problem and that the whole system is reliable when employed for a series of failures.
{"title":"Towards a Smart Fault Tolerant Indoor Localization System Through Recurrent Neural Networks","authors":"E. Carvalho, Bruno V. Ferreira, G. P. R. Filho, P. Gomes, G. Freitas, P. A. Vargas, J. Ueyama, G. Pessin","doi":"10.1109/IJCNN.2019.8852007","DOIUrl":"https://doi.org/10.1109/IJCNN.2019.8852007","url":null,"abstract":"This paper proposes a fault-tolerant indoor localization system that employs Recurrent Neural Networks (RNNs) for the localization task. A decision module is designed to detect failures and this is responsible for the allocation of RNNs that are suitable for each situation. As well as the fault-tolerant system, several architectures and models for RNNs are exploited in the system: Gated Recurrent Unit (GRU), Long Short-Term Memory (LSTM) and Simple RNN. The system uses as inputs a collection of Wi-Fi Received Signal Strength Indication (RSSI) signals, and the RNN classifies the position of an agent on the basis of this collection. A fault-tolerant mechanism has been designed to handle two types of failures: (i) momentary failure, and (ii) permanent failure. The results show that the RNNs are suitable for tackling the problem and that the whole system is reliable when employed for a series of failures.","PeriodicalId":222797,"journal":{"name":"2019 International Joint Conference on Neural Networks (IJCNN)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132642252","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}
Pub Date : 2019-09-30DOI: 10.1109/IJCNN.2019.8852279
Federico Corradi, S. Pande, J. Stuijt, Ning Qiao, S. Schaafsma, G. Indiveri, F. Catthoor
Heart activity can be monitored by means of ElectroCardioGram (ECG) measure which is widely used to detect heart diseases due to its non-invasive nature. Trained cardiologists can detect anomalies by visual inspecting recordings of the ECG signals. However, arrhythmias occur intermittently especially in early stages and therefore they can be missed in routine check recordings. We propose a hardware setup that enables the always-on monitoring of ECG signals into wearables. The system exploits a fully event-driven approach for carrying arrhythmia detection and classification employing a bio-inspired spiking neural network. The two staged Spiking Neural Network (SNN) topology comprises a recurrent network of spiking neurons whose output is classified by a cluster of Leaky integrate-and-fire (LIF) neurons that have been supervisely trained to distinguish 17 types of cardiac patterns. We introduce a method for compressing ECG signals into a stream of asynchronous digital events that are used to stimulate the recurrent SNN. Using ablative analysis, we demonstrate the impact of the recurrent SNN and we show an overall classification accuracy of 95% on the PhysioNet Arrhythmia Database provided by the Massachusetts Institute of Technology and Beth Israel Hospital (MIT/BIH). The proposed system has been implemented on an event-driven mixed-signal analog/digital neuromorphic processor. This work contributes to the realization of an energy-efficient, wearable, and accurate multi-class ECG classification system.
{"title":"ECG-based Heartbeat Classification in Neuromorphic Hardware","authors":"Federico Corradi, S. Pande, J. Stuijt, Ning Qiao, S. Schaafsma, G. Indiveri, F. Catthoor","doi":"10.1109/IJCNN.2019.8852279","DOIUrl":"https://doi.org/10.1109/IJCNN.2019.8852279","url":null,"abstract":"Heart activity can be monitored by means of ElectroCardioGram (ECG) measure which is widely used to detect heart diseases due to its non-invasive nature. Trained cardiologists can detect anomalies by visual inspecting recordings of the ECG signals. However, arrhythmias occur intermittently especially in early stages and therefore they can be missed in routine check recordings. We propose a hardware setup that enables the always-on monitoring of ECG signals into wearables. The system exploits a fully event-driven approach for carrying arrhythmia detection and classification employing a bio-inspired spiking neural network. The two staged Spiking Neural Network (SNN) topology comprises a recurrent network of spiking neurons whose output is classified by a cluster of Leaky integrate-and-fire (LIF) neurons that have been supervisely trained to distinguish 17 types of cardiac patterns. We introduce a method for compressing ECG signals into a stream of asynchronous digital events that are used to stimulate the recurrent SNN. Using ablative analysis, we demonstrate the impact of the recurrent SNN and we show an overall classification accuracy of 95% on the PhysioNet Arrhythmia Database provided by the Massachusetts Institute of Technology and Beth Israel Hospital (MIT/BIH). The proposed system has been implemented on an event-driven mixed-signal analog/digital neuromorphic processor. This work contributes to the realization of an energy-efficient, wearable, and accurate multi-class ECG classification system.","PeriodicalId":222797,"journal":{"name":"2019 International Joint Conference on Neural Networks (IJCNN)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124397718","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}
Pub Date : 2019-09-30DOI: 10.1109/IJCNN.2019.8852140
Amna Dridi, M. Gaber, R. Azad, Jagdev Bhogal
Research in analysis of big scholarly data has increased in the recent past and it aims to understand research dynamics and forecast research trends. The ultimate objective in this research is to design and implement novel and scalable methods for extracting knowledge and computational history. While citations are highly used to identify emerging/rising research topics, they can take months or even years to stabilise enough to reveal research trends. Consequently, it is necessary to develop faster yet accurate methods for trend analysis and computational history that dig into content and semantics of an article. Therefore, this paper aims to conduct a fine-grained content analysis of scientific corpora from the domain of Machine Learning. This analysis uses DeepHist, a deep learning-based computational history approach; the approach relies on a dynamic word embedding that aims to represent words with low-dimensional vectors computed by deep neural networks. The scientific corpora come from 5991 publications from Neural Information Processing Systems (NIPS) conference between 1987 and 2015 which are divided into six 5-year timespans. The analysis of these corpora generates visualisations produced by applying t-distributed stochastic neighbor embedding (t-SNE) for dimensionality reduction. The qualitative and quantitative study reported here reveals the evolution of the prominent Machine Learning keywords; this evolution supports the popularity of current research topics in the field. This support is evident given how well the popularity of the detected keywords correlates with the citation counts received by their corresponding papers: Spearman’s positive correlation is 100%. With such a strong result, this work evidences the utility of deep learning techniques for determining the computational history of science.
{"title":"DeepHist: Towards a Deep Learning-based Computational History of Trends in the NIPS","authors":"Amna Dridi, M. Gaber, R. Azad, Jagdev Bhogal","doi":"10.1109/IJCNN.2019.8852140","DOIUrl":"https://doi.org/10.1109/IJCNN.2019.8852140","url":null,"abstract":"Research in analysis of big scholarly data has increased in the recent past and it aims to understand research dynamics and forecast research trends. The ultimate objective in this research is to design and implement novel and scalable methods for extracting knowledge and computational history. While citations are highly used to identify emerging/rising research topics, they can take months or even years to stabilise enough to reveal research trends. Consequently, it is necessary to develop faster yet accurate methods for trend analysis and computational history that dig into content and semantics of an article. Therefore, this paper aims to conduct a fine-grained content analysis of scientific corpora from the domain of Machine Learning. This analysis uses DeepHist, a deep learning-based computational history approach; the approach relies on a dynamic word embedding that aims to represent words with low-dimensional vectors computed by deep neural networks. The scientific corpora come from 5991 publications from Neural Information Processing Systems (NIPS) conference between 1987 and 2015 which are divided into six 5-year timespans. The analysis of these corpora generates visualisations produced by applying t-distributed stochastic neighbor embedding (t-SNE) for dimensionality reduction. The qualitative and quantitative study reported here reveals the evolution of the prominent Machine Learning keywords; this evolution supports the popularity of current research topics in the field. This support is evident given how well the popularity of the detected keywords correlates with the citation counts received by their corresponding papers: Spearman’s positive correlation is 100%. With such a strong result, this work evidences the utility of deep learning techniques for determining the computational history of science.","PeriodicalId":222797,"journal":{"name":"2019 International Joint Conference on Neural Networks (IJCNN)","volume":"213 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131494718","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}
Pub Date : 2019-09-30DOI: 10.1109/IJCNN.2019.8851919
W. Chin, N. Kubota, C. Loo, Zhaojie Ju, Honghai Liu
Biological information inspires the advancement of a navigational mechanism for autonomous robots to help people explore and map real-world environments. However, the robot’s ability to constantly acquire environmental information in real-world, dynamic environments has remained a challenge for many years. In this paper, we propose a self-organizing adaptive recurrent incremental network that models human episodic memory to learn spatiotemporal representations from novel sensory data. The proposed method termed as SOARIN consists of two main learning process that is active learning and episodic memory playback. For active learning (robot exploration), SOARIN quickly learns and adapts incoming novel sensory data as episodic neurons via competitive Hebbian Learning. Episodic neurons are connecting with each other and gradually forms a spatial map that can be used for robot localization. Episodic memory playback is triggered whenever the robot is in an inactive mode (charging or hibernating). During playback, SOARIN gradually integrates knowledge and experience into more consolidate spatial map structures that can overcome the catastrophic forgetting. The proposed method is analyzed and evaluated in term of map learning and localization through a series of real robot experiments in real-world indoor environments.
{"title":"Spatial Map Learning with Self-Organizing Adaptive Recurrent Incremental Network","authors":"W. Chin, N. Kubota, C. Loo, Zhaojie Ju, Honghai Liu","doi":"10.1109/IJCNN.2019.8851919","DOIUrl":"https://doi.org/10.1109/IJCNN.2019.8851919","url":null,"abstract":"Biological information inspires the advancement of a navigational mechanism for autonomous robots to help people explore and map real-world environments. However, the robot’s ability to constantly acquire environmental information in real-world, dynamic environments has remained a challenge for many years. In this paper, we propose a self-organizing adaptive recurrent incremental network that models human episodic memory to learn spatiotemporal representations from novel sensory data. The proposed method termed as SOARIN consists of two main learning process that is active learning and episodic memory playback. For active learning (robot exploration), SOARIN quickly learns and adapts incoming novel sensory data as episodic neurons via competitive Hebbian Learning. Episodic neurons are connecting with each other and gradually forms a spatial map that can be used for robot localization. Episodic memory playback is triggered whenever the robot is in an inactive mode (charging or hibernating). During playback, SOARIN gradually integrates knowledge and experience into more consolidate spatial map structures that can overcome the catastrophic forgetting. The proposed method is analyzed and evaluated in term of map learning and localization through a series of real robot experiments in real-world indoor environments.","PeriodicalId":222797,"journal":{"name":"2019 International Joint Conference on Neural Networks (IJCNN)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114798497","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}
Pub Date : 2019-09-30DOI: 10.1109/IJCNN.2019.8851842
E. Soares, P. Angelov, Bruno Costa, Marcos Castro
This paper presents an actively semi-supervised multi-layer neuro-fuzzy modeling method, ASSDRB, to classify different lighting conditions for driving scenes. ASSDRB is composed of a massively parallel ensemble of AnYa type 0-order fuzzy rules. It uses a recursive learning algorithm to update its structure when new data items are provided and, therefore, is able to cope with nonstationarities. Different lighting conditions for driving situations are considered in the analysis, which is used by self-driving cars as a safety mechanism. Differently from mainstream Deep Neural Networks approaches, the ASSDRB is able to learn from unseen data. Experiments on different lighting conditions for driving scenes, demonstrated that the deep neuro-fuzzy modeling is an efficient framework for these challenging classification tasks. Classification accuracy is higher than those produced by alternative machine learning methods. The number of algebraic calculations for the present method are significantly smaller and, therefore, the method is significantly faster than common Deep Neural Networks approaches. Moreover, DRB produced transparent AnYa fuzzy rules, which are human interpretable.
{"title":"Actively Semi-Supervised Deep Rule-based Classifier Applied to Adverse Driving Scenarios","authors":"E. Soares, P. Angelov, Bruno Costa, Marcos Castro","doi":"10.1109/IJCNN.2019.8851842","DOIUrl":"https://doi.org/10.1109/IJCNN.2019.8851842","url":null,"abstract":"This paper presents an actively semi-supervised multi-layer neuro-fuzzy modeling method, ASSDRB, to classify different lighting conditions for driving scenes. ASSDRB is composed of a massively parallel ensemble of AnYa type 0-order fuzzy rules. It uses a recursive learning algorithm to update its structure when new data items are provided and, therefore, is able to cope with nonstationarities. Different lighting conditions for driving situations are considered in the analysis, which is used by self-driving cars as a safety mechanism. Differently from mainstream Deep Neural Networks approaches, the ASSDRB is able to learn from unseen data. Experiments on different lighting conditions for driving scenes, demonstrated that the deep neuro-fuzzy modeling is an efficient framework for these challenging classification tasks. Classification accuracy is higher than those produced by alternative machine learning methods. The number of algebraic calculations for the present method are significantly smaller and, therefore, the method is significantly faster than common Deep Neural Networks approaches. Moreover, DRB produced transparent AnYa fuzzy rules, which are human interpretable.","PeriodicalId":222797,"journal":{"name":"2019 International Joint Conference on Neural Networks (IJCNN)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128793555","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}
Pub Date : 2019-07-19DOI: 10.1109/IJCNN.2019.8852303
C. Singh, V. Gangwar, H. Singh, Karan Narain, A. Majumder, Swagat Kumar
Automatic recognition of text, such as a batch code printed on a box placed on a moving conveyor belt, is still a challenging problem. This paper proposes an end-to-end character recognition technique while addressing the major challenges encountered in a real environment, such as motion blur in the acquired images, slanted or oriented characters, creased batch codes due to wear and tear of boxes, variations in label formats, and variations in printing styles. The major contribution of this work lies in development of three sequential modules: text localization using Connectionist Text Proposal Network(CTPN), character detection and character recognition using a modified version of the capsule network (CapsNet). In contrast to CapsNet, where only a standard single convolution is used, the proposed method uses a series of feature blocks, making it a deep CapsNet which is later proven to generate more comprehensive and better separable feature vectors over its counterpart. The feature generation module is further enhanced by setting a smaller kernel size than CapsNet. The proposed system is validated on a real-world box / packet dataset generated in a retail manufacturing industry. The proposed recognition network architecture is also validated on a standard public dataset (ICDAR 2013). The comparative results are presented with statistical analysis in the experimental results section.
文本的自动识别，例如打印在移动传送带上的盒子上的批处理代码，仍然是一个具有挑战性的问题。本文提出了一种端到端字符识别技术，同时解决了在真实环境中遇到的主要挑战，例如获取的图像中的运动模糊，倾斜或定向字符，由于盒子磨损而产生的批码折痕，标签格式的变化以及打印样式的变化。这项工作的主要贡献在于开发了三个连续的模块:使用Connectionist text Proposal Network(CTPN)的文本定位，使用改进版本的capsule Network(CapsNet)的字符检测和字符识别。与仅使用标准单个卷积的CapsNet相比，所提出的方法使用一系列特征块，使其成为一个深度CapsNet，后来被证明可以生成比其对应的更全面和更好的可分离特征向量。通过设置比CapsNet更小的内核大小，功能生成模块得到了进一步增强。提出的系统在零售制造业生成的真实盒子/包数据集上进行了验证。提出的识别网络架构也在标准公共数据集(ICDAR 2013)上进行了验证。实验结果部分给出了对比结果并进行了统计分析。
{"title":"Deep Capsule Network based Automatic Batch Code Identification Pipeline for a Real-life Industrial Application","authors":"C. Singh, V. Gangwar, H. Singh, Karan Narain, A. Majumder, Swagat Kumar","doi":"10.1109/IJCNN.2019.8852303","DOIUrl":"https://doi.org/10.1109/IJCNN.2019.8852303","url":null,"abstract":"Automatic recognition of text, such as a batch code printed on a box placed on a moving conveyor belt, is still a challenging problem. This paper proposes an end-to-end character recognition technique while addressing the major challenges encountered in a real environment, such as motion blur in the acquired images, slanted or oriented characters, creased batch codes due to wear and tear of boxes, variations in label formats, and variations in printing styles. The major contribution of this work lies in development of three sequential modules: text localization using Connectionist Text Proposal Network(CTPN), character detection and character recognition using a modified version of the capsule network (CapsNet). In contrast to CapsNet, where only a standard single convolution is used, the proposed method uses a series of feature blocks, making it a deep CapsNet which is later proven to generate more comprehensive and better separable feature vectors over its counterpart. The feature generation module is further enhanced by setting a smaller kernel size than CapsNet. The proposed system is validated on a real-world box / packet dataset generated in a retail manufacturing industry. The proposed recognition network architecture is also validated on a standard public dataset (ICDAR 2013). The comparative results are presented with statistical analysis in the experimental results section.","PeriodicalId":222797,"journal":{"name":"2019 International Joint Conference on Neural Networks (IJCNN)","volume":"34 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123354645","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}
Pub Date : 2019-07-15DOI: 10.1109/IJCNN.2019.8852206
Sahar Arshi, Li Zhang, Rebecca Strachan
Photovoltaic (PV) systems use the sunlight and convert it to electrical power. It is predicted that by 2023, 371,000 PV installations will be embedded in power networks in the UK. This may increase the risk of voltage rise which has adverse impacts on the power network. The balance maintenance is important for high security of the physical electrical systems and the operation economy. Therefore, the prediction of the output of PV systems is of great importance. The output of a PV system highly depends on local environmental conditions. These include sun radiation, temperature, and humidity. In this research, the importance of various weather factors are studied. The weather attributes are subsequently employed for the prediction of the solar panel power generation from a time-series database. Long-Short Term Memory networks are employed for obtaining the dependencies between various elements of the weather conditions and the PV energy metrics. Evaluation results indicate the efficiency of the deep networks for energy generation prediction.
{"title":"Prediction Using LSTM Networks","authors":"Sahar Arshi, Li Zhang, Rebecca Strachan","doi":"10.1109/IJCNN.2019.8852206","DOIUrl":"https://doi.org/10.1109/IJCNN.2019.8852206","url":null,"abstract":"Photovoltaic (PV) systems use the sunlight and convert it to electrical power. It is predicted that by 2023, 371,000 PV installations will be embedded in power networks in the UK. This may increase the risk of voltage rise which has adverse impacts on the power network. The balance maintenance is important for high security of the physical electrical systems and the operation economy. Therefore, the prediction of the output of PV systems is of great importance. The output of a PV system highly depends on local environmental conditions. These include sun radiation, temperature, and humidity. In this research, the importance of various weather factors are studied. The weather attributes are subsequently employed for the prediction of the solar panel power generation from a time-series database. Long-Short Term Memory networks are employed for obtaining the dependencies between various elements of the weather conditions and the PV energy metrics. Evaluation results indicate the efficiency of the deep networks for energy generation prediction.","PeriodicalId":222797,"journal":{"name":"2019 International Joint Conference on Neural Networks (IJCNN)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116964124","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}
Pub Date : 2019-07-14DOI: 10.1109/IJCNN.2019.8851939
Fayeem Aziz, S. Chalup
Manifold learning can only be successful if enough data is available. If the data is too sparse, the geometrical and topological structure of the manifold extracted from the data cannot be recognised and the manifold collapses. In this paper we used data from a simulated two-dimensional double pendulum and tested how well several manifold learning methods could extract the expected manifold, a two-dimensional torus. The experiments were repeated while the data was downsampled in several ways to test the robustness of the different manifold learning methods. We also developed a neural network-based deep autoencoder for manifold learning and demonstrated that it performed in most of our test cases similarly or better than traditional methods such as principal component analysis and isomap.
{"title":"Testing the Robustness of Manifold Learning on Examples of Thinned-Out Data","authors":"Fayeem Aziz, S. Chalup","doi":"10.1109/IJCNN.2019.8851939","DOIUrl":"https://doi.org/10.1109/IJCNN.2019.8851939","url":null,"abstract":"Manifold learning can only be successful if enough data is available. If the data is too sparse, the geometrical and topological structure of the manifold extracted from the data cannot be recognised and the manifold collapses. In this paper we used data from a simulated two-dimensional double pendulum and tested how well several manifold learning methods could extract the expected manifold, a two-dimensional torus. The experiments were repeated while the data was downsampled in several ways to test the robustness of the different manifold learning methods. We also developed a neural network-based deep autoencoder for manifold learning and demonstrated that it performed in most of our test cases similarly or better than traditional methods such as principal component analysis and isomap.","PeriodicalId":222797,"journal":{"name":"2019 International Joint Conference on Neural Networks (IJCNN)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121830139","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}
Pub Date : 2019-07-14DOI: 10.1109/IJCNN.2019.8851797
Mehdi Abadi, S. Jovanovic, K. Khalifa, S. Weber, M. H. Bedoui
In this work, a scalable and adaptable hardware SOM architecture allowing to execute multiple applications in parallel is presented. The proposed architecture allows to use simultaneously multiple SOM structures with different parameters in order to satisfy multiple applications with different needs. The application switching is done within a clock cycle at the neuron’s level at run time only by analyzing the received input data. The proposed architecture was tested and validated in an image quantization experiment where 6 quantization applications with different parameters (from 6 × 6 to 15 × 15 SOMs with inputs varying from 3 to 12 elements) were performed simultaneously.
{"title":"A Multi-Application, Scalable and Adaptable Hardware SOM Architecture","authors":"Mehdi Abadi, S. Jovanovic, K. Khalifa, S. Weber, M. H. Bedoui","doi":"10.1109/IJCNN.2019.8851797","DOIUrl":"https://doi.org/10.1109/IJCNN.2019.8851797","url":null,"abstract":"In this work, a scalable and adaptable hardware SOM architecture allowing to execute multiple applications in parallel is presented. The proposed architecture allows to use simultaneously multiple SOM structures with different parameters in order to satisfy multiple applications with different needs. The application switching is done within a clock cycle at the neuron’s level at run time only by analyzing the received input data. The proposed architecture was tested and validated in an image quantization experiment where 6 quantization applications with different parameters (from 6 × 6 to 15 × 15 SOMs with inputs varying from 3 to 12 elements) were performed simultaneously.","PeriodicalId":222797,"journal":{"name":"2019 International Joint Conference on Neural Networks (IJCNN)","volume":"93 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127341189","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：