Pub Date : 2019-03-01DOI: 10.1109/CISS.2019.8693048
Wondimu K. Zegeye, F. Moazzami
The deployment of computing devices and machines that constitute the Internet of Things (IoT) had dramatically increased in this decade. To reduce the cost of pre-deployment effort, an already existing Wi-Fi infrastructure (Access Point) can be used to grant these devices access to Internet connectivity. This paper presents the use of standard tunneled based EAP (TEAP) protocol for authenticating IoT devices from the cloud using an EAP-AKA’ inner method EAP. It mainly focuses on the deployment architecture, security against MiTM attack by exploring different RFCs associated with those protocols.
{"title":"Authentication of IoT Devices for WiFi Connectivity from the Cloud","authors":"Wondimu K. Zegeye, F. Moazzami","doi":"10.1109/CISS.2019.8693048","DOIUrl":"https://doi.org/10.1109/CISS.2019.8693048","url":null,"abstract":"The deployment of computing devices and machines that constitute the Internet of Things (IoT) had dramatically increased in this decade. To reduce the cost of pre-deployment effort, an already existing Wi-Fi infrastructure (Access Point) can be used to grant these devices access to Internet connectivity. This paper presents the use of standard tunneled based EAP (TEAP) protocol for authenticating IoT devices from the cloud using an EAP-AKA’ inner method EAP. It mainly focuses on the deployment architecture, security against MiTM attack by exploring different RFCs associated with those protocols.","PeriodicalId":123696,"journal":{"name":"2019 53rd Annual Conference on Information Sciences and Systems (CISS)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132348022","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 : 2019-03-01DOI: 10.1109/CISS.2019.8693028
Sima Azizi, A. Komaee
Symmetric cryptography relies on pairs of identical secret keys shared by the legitimate communicating parties. To implement a symmetric-key algorithm for cryptography, a major concern is to develop secure methods for distribution of the secret key. In securing the wireless fading channels by symmetric-key algorithms, the physical layer properties of the channel can be exploited for distribution of the secret keys. In this approach, the channel state provides a common randomness which is shared by the legitimate users but is mostly unknown to an eavesdropper. by means of signal processing techniques, this common randomness is extracted into random secret keys. This paper establishes an information theoretic upper bound on the rate at which the secret keys can be extracted. Instead of the conventional approach that relies on mathematical models for the wireless channel, this paper adopts an experimental approach to estimate this bound from empirical data. A set of signal processing techniques is developed here to numerically estimate this bound for a pair of received signal strength (RSS) recorded by indoor commercial radios.
{"title":"Empirical Bounds on the Rate of Secret Bits Extracted from Received Signal Strength","authors":"Sima Azizi, A. Komaee","doi":"10.1109/CISS.2019.8693028","DOIUrl":"https://doi.org/10.1109/CISS.2019.8693028","url":null,"abstract":"Symmetric cryptography relies on pairs of identical secret keys shared by the legitimate communicating parties. To implement a symmetric-key algorithm for cryptography, a major concern is to develop secure methods for distribution of the secret key. In securing the wireless fading channels by symmetric-key algorithms, the physical layer properties of the channel can be exploited for distribution of the secret keys. In this approach, the channel state provides a common randomness which is shared by the legitimate users but is mostly unknown to an eavesdropper. by means of signal processing techniques, this common randomness is extracted into random secret keys. This paper establishes an information theoretic upper bound on the rate at which the secret keys can be extracted. Instead of the conventional approach that relies on mathematical models for the wireless channel, this paper adopts an experimental approach to estimate this bound from empirical data. A set of signal processing techniques is developed here to numerically estimate this bound for a pair of received signal strength (RSS) recorded by indoor commercial radios.","PeriodicalId":123696,"journal":{"name":"2019 53rd Annual Conference on Information Sciences and Systems (CISS)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121068598","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 : 2019-03-01DOI: 10.1109/CISS.2019.8693024
Mariem Ben Fadhel, K. Nyarko
Anomaly detection is the process of finding data points that deviate from a baseline. In a real-life setting, anomalies are usually unknown or extremely rare. Moreover, the detection must be accomplished in a timely manner or the risk of corrupting the system might grow exponentially. In this work, we propose a two level framework for detecting anomalies in sequences of discrete elements. First, we assess whether we can obtain enough information from the statistics collected from the discriminator’s layers to discriminate between out of distribution and in distribution samples. We then build an unsupervised anomaly detection module based on these statistics. As to augment the data and keep track of classes of known data, we lean toward a semi-supervised adversarial learning applied to discrete elements.
{"title":"GAN Augmented Text Anomaly Detection with Sequences of Deep Statistics","authors":"Mariem Ben Fadhel, K. Nyarko","doi":"10.1109/CISS.2019.8693024","DOIUrl":"https://doi.org/10.1109/CISS.2019.8693024","url":null,"abstract":"Anomaly detection is the process of finding data points that deviate from a baseline. In a real-life setting, anomalies are usually unknown or extremely rare. Moreover, the detection must be accomplished in a timely manner or the risk of corrupting the system might grow exponentially. In this work, we propose a two level framework for detecting anomalies in sequences of discrete elements. First, we assess whether we can obtain enough information from the statistics collected from the discriminator’s layers to discriminate between out of distribution and in distribution samples. We then build an unsupervised anomaly detection module based on these statistics. As to augment the data and keep track of classes of known data, we lean toward a semi-supervised adversarial learning applied to discrete elements.","PeriodicalId":123696,"journal":{"name":"2019 53rd Annual Conference on Information Sciences and Systems (CISS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129309374","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 : 2019-03-01DOI: 10.1109/CISS.2019.8692879
Peter O. Taiwo, A. Cole-Rhodes
In this work we explore the use of adaptive beamforming for millimeter wave (mm-wave) communications. We apply a blind adaptive algorithm to blocks of signal data transmitted using 16-QAM modulation. We consider uplink transmission for both a single user and for multiple users, each equipped with a single antenna. The receiver is equipped with a uniform linear array (ULA) antenna. We estimate the effective channel and measure the performance of a blind millimeter wave combiner applied at the base station to received signal blocks, which have been corrupted by AWGN and inter-symbol interference (ISI). This is done by measuring error rates over varying SNR, and varying numbers of antenna at the base station.
{"title":"Adaptive Beamforming for Multiple-Access Millimeter Wave Communications : Invited Presentation","authors":"Peter O. Taiwo, A. Cole-Rhodes","doi":"10.1109/CISS.2019.8692879","DOIUrl":"https://doi.org/10.1109/CISS.2019.8692879","url":null,"abstract":"In this work we explore the use of adaptive beamforming for millimeter wave (mm-wave) communications. We apply a blind adaptive algorithm to blocks of signal data transmitted using 16-QAM modulation. We consider uplink transmission for both a single user and for multiple users, each equipped with a single antenna. The receiver is equipped with a uniform linear array (ULA) antenna. We estimate the effective channel and measure the performance of a blind millimeter wave combiner applied at the base station to received signal blocks, which have been corrupted by AWGN and inter-symbol interference (ISI). This is done by measuring error rates over varying SNR, and varying numbers of antenna at the base station.","PeriodicalId":123696,"journal":{"name":"2019 53rd Annual Conference on Information Sciences and Systems (CISS)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127957838","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 : 2019-03-01DOI: 10.1109/CISS.2019.8693021
Pan Zhong, Zhengdao Wang
The group convolution and representation theory give a strong support for generalized convolutional neural network. The generalized convolutional neural network (G-CNN) has been applied to learning problems and achieved the state-of-art performance. But a theoretical support for details of network architecture design is still lacking. In this work, we first analyze the necessary and sufficient condition for a neural network to be group equivariant when the group acts on the sub-domain of input/output. We then analyze the multiple equivariance case. After that, we show that the generalized convolution mapping to a quotient space is a projection of the image of a generalized convolution which maps to the maximum quotient space. This can be used to obtain guidelines for choosing the feature size of hidden layer.
{"title":"Characterization and Design of Generalized Convolutional Neural Network","authors":"Pan Zhong, Zhengdao Wang","doi":"10.1109/CISS.2019.8693021","DOIUrl":"https://doi.org/10.1109/CISS.2019.8693021","url":null,"abstract":"The group convolution and representation theory give a strong support for generalized convolutional neural network. The generalized convolutional neural network (G-CNN) has been applied to learning problems and achieved the state-of-art performance. But a theoretical support for details of network architecture design is still lacking. In this work, we first analyze the necessary and sufficient condition for a neural network to be group equivariant when the group acts on the sub-domain of input/output. We then analyze the multiple equivariance case. After that, we show that the generalized convolution mapping to a quotient space is a projection of the image of a generalized convolution which maps to the maximum quotient space. This can be used to obtain guidelines for choosing the feature size of hidden layer.","PeriodicalId":123696,"journal":{"name":"2019 53rd Annual Conference on Information Sciences and Systems (CISS)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125942450","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 : 2019-03-01DOI: 10.1109/CISS.2019.8693042
Sushanth G. Sathyanarayana, Bo Ning, Song Hu, J. Hossack
Dictionary learning is an unsupervised learning method to abstract image data into a set of learned basis vectors. In prior work, the efficacy of the K-SVD dictionary learning algorithm in suppressing reverberation in volumetric photoacoustic microscopy (PAM) data was demonstrated. In this work, we compare the K-SVD algorithm against the method of optimal directions (MOD). The generalization error and reverberation suppression performance of the two algorithms were compared. The K-SVD was found to have a lower average generalization error (5.69x104 ±9.09x103 (a.u.)) when compared to the MOD (8.27x104 ±1.33x104 (a.u.)) for identical training data, initialization, sparsity (3 atoms per A-line) and number of iterations (5). Both algorithms were observed to suppress the reverberation to a similar extent (18.8 ± 1.12 dB for the K-SVD and 18.3 ± 1.2 dB for the MOD). Our data show that irrespective of the method used, sparse dictionary learning can significantly suppress reverberations in PAM.
{"title":"Comparison of dictionary learning methods for reverberation suppression in photoacoustic microscopy : Invited presentation","authors":"Sushanth G. Sathyanarayana, Bo Ning, Song Hu, J. Hossack","doi":"10.1109/CISS.2019.8693042","DOIUrl":"https://doi.org/10.1109/CISS.2019.8693042","url":null,"abstract":"Dictionary learning is an unsupervised learning method to abstract image data into a set of learned basis vectors. In prior work, the efficacy of the K-SVD dictionary learning algorithm in suppressing reverberation in volumetric photoacoustic microscopy (PAM) data was demonstrated. In this work, we compare the K-SVD algorithm against the method of optimal directions (MOD). The generalization error and reverberation suppression performance of the two algorithms were compared. The K-SVD was found to have a lower average generalization error (5.69x104 ±9.09x103 (a.u.)) when compared to the MOD (8.27x104 ±1.33x104 (a.u.)) for identical training data, initialization, sparsity (3 atoms per A-line) and number of iterations (5). Both algorithms were observed to suppress the reverberation to a similar extent (18.8 ± 1.12 dB for the K-SVD and 18.3 ± 1.2 dB for the MOD). Our data show that irrespective of the method used, sparse dictionary learning can significantly suppress reverberations in PAM.","PeriodicalId":123696,"journal":{"name":"2019 53rd Annual Conference on Information Sciences and Systems (CISS)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114316716","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 : 2019-03-01DOI: 10.1109/CISS.2019.8692855
Q. H. Cap, Hiroki Tani, H. Uga, S. Kagiwada, H. Iyatomi
Automated plant diagnosis using images taken from a distance is often insufficient in resolution and degrades diagnostic accuracy since the important external characteristics of symptoms are lost. In this paper, we first propose an effective preprocessing method for improving the performance of automated plant disease diagnosis systems using super-resolution techniques. We investigate the efficiency of two different super-resolution methods by comparing the disease diagnostic performance on the practical original high-resolution, low-resolution, and super-resolved cucumber images. Our method generates super-resolved images that look very close to natural images with 4 × upscaling factors and is capable of recovering the lost detailed symptoms, largely boosting the diagnostic performance. Our model improves the disease classification accuracy by 26.9% over the bicubic interpolation method of 65.6% and shows a small gap (3% lower) between the original result of 95.5%.
{"title":"Super-Resolution for Practical Automated Plant Disease Diagnosis System","authors":"Q. H. Cap, Hiroki Tani, H. Uga, S. Kagiwada, H. Iyatomi","doi":"10.1109/CISS.2019.8692855","DOIUrl":"https://doi.org/10.1109/CISS.2019.8692855","url":null,"abstract":"Automated plant diagnosis using images taken from a distance is often insufficient in resolution and degrades diagnostic accuracy since the important external characteristics of symptoms are lost. In this paper, we first propose an effective preprocessing method for improving the performance of automated plant disease diagnosis systems using super-resolution techniques. We investigate the efficiency of two different super-resolution methods by comparing the disease diagnostic performance on the practical original high-resolution, low-resolution, and super-resolved cucumber images. Our method generates super-resolved images that look very close to natural images with 4 × upscaling factors and is capable of recovering the lost detailed symptoms, largely boosting the diagnostic performance. Our model improves the disease classification accuracy by 26.9% over the bicubic interpolation method of 65.6% and shows a small gap (3% lower) between the original result of 95.5%.","PeriodicalId":123696,"journal":{"name":"2019 53rd Annual Conference on Information Sciences and Systems (CISS)","volume":"1 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133304692","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 : 2019-03-01DOI: 10.1109/CISS.2019.8692919
H. B. Kassa, K. Kornegay, Ebrima N. Ceesay
This paper proposes an adaptive algorithm to maximize energy efficiency in cellular network considering a dynamic user clustering technique. First, a base station (BS) sleeping algorithm is designed, which minimizes the energy consumption to almost more than half. Then a Linear Radius User Clustering algorithm is modeled. Using feedback channel state information to the base station, the algorithm varies the mobile cell radius adaptively to minimize a total energy consumption of overall cellular network based on the threshold user density. The minimum distance where a Mobile Station can get a signal from the base station without a significant effect on human health can be located. Since the Base Station with modern scanner installed on its transmitter part can scan 390 times per second, the time scale to marginalize users from the coverage under threshold densities is in milliseconds. As a result, there is no significant effect on quality of services when the cell coverage is zoomed in/out periodically. Numerical results show that the proposed algorithm can considerably reduce energy consumption compared with the cases where a base station is always turned on with constant maximum transmit power.
{"title":"Energy Efficient Cellular Network User Clustering Using Linear Radius Algorithm.","authors":"H. B. Kassa, K. Kornegay, Ebrima N. Ceesay","doi":"10.1109/CISS.2019.8692919","DOIUrl":"https://doi.org/10.1109/CISS.2019.8692919","url":null,"abstract":"This paper proposes an adaptive algorithm to maximize energy efficiency in cellular network considering a dynamic user clustering technique. First, a base station (BS) sleeping algorithm is designed, which minimizes the energy consumption to almost more than half. Then a Linear Radius User Clustering algorithm is modeled. Using feedback channel state information to the base station, the algorithm varies the mobile cell radius adaptively to minimize a total energy consumption of overall cellular network based on the threshold user density. The minimum distance where a Mobile Station can get a signal from the base station without a significant effect on human health can be located. Since the Base Station with modern scanner installed on its transmitter part can scan 390 times per second, the time scale to marginalize users from the coverage under threshold densities is in milliseconds. As a result, there is no significant effect on quality of services when the cell coverage is zoomed in/out periodically. Numerical results show that the proposed algorithm can considerably reduce energy consumption compared with the cases where a base station is always turned on with constant maximum transmit power.","PeriodicalId":123696,"journal":{"name":"2019 53rd Annual Conference on Information Sciences and Systems (CISS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130862091","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 : 2019-03-01DOI: 10.1109/CISS.2019.8693034
Jay Jemal, K. Kornegay
As Blockchain technology become more understood in recent years and its capability to solve enterprise business use cases become evident, technologist have been exploring Blockchain technology to solve use cases that have been daunting industries for years. Unlike existing technologies, one of the key features of blockchain technology is its unparalleled capability to provide, traceability, accountability and immutable records that can be accessed at any point in time. One application area of interest for blockchain is securing heterogenous networks. This paper explores the security challenges in a heterogonous network of IoT devices and whether blockchain can be a viable solution. Using an experimental approach, we explore the possibility of using blockchain technology to secure IoT devices, validate IoT device transactions, and establish a chain of trust to secure an IoT device mesh network, as well as investigate the plausibility of using immutable transactions for forensic analysis.
{"title":"Security Assessment of Blockchains in Heterogenous IoT Networks : Invited Presentation","authors":"Jay Jemal, K. Kornegay","doi":"10.1109/CISS.2019.8693034","DOIUrl":"https://doi.org/10.1109/CISS.2019.8693034","url":null,"abstract":"As Blockchain technology become more understood in recent years and its capability to solve enterprise business use cases become evident, technologist have been exploring Blockchain technology to solve use cases that have been daunting industries for years. Unlike existing technologies, one of the key features of blockchain technology is its unparalleled capability to provide, traceability, accountability and immutable records that can be accessed at any point in time. One application area of interest for blockchain is securing heterogenous networks. This paper explores the security challenges in a heterogonous network of IoT devices and whether blockchain can be a viable solution. Using an experimental approach, we explore the possibility of using blockchain technology to secure IoT devices, validate IoT device transactions, and establish a chain of trust to secure an IoT device mesh network, as well as investigate the plausibility of using immutable transactions for forensic analysis.","PeriodicalId":123696,"journal":{"name":"2019 53rd Annual Conference on Information Sciences and Systems (CISS)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127625249","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 : 2019-03-01DOI: 10.1109/CISS.2019.8692932
Hossein Valavi, P. Ramadge
Low-rank matrix factorization can reveal fundamental structure in data. For example, joint-PCA on multi-datasets can find a joint, lower-dimensional representation of the data. Recently other similar matrix factorization methods have been introduced for multi-dataset analysis, e.g., the shared response model (SRM) and hyperalignment (HA). We provide a comparison of these methods with joint-PCA that highlights similarities and differences. Necessary and sufficient conditions under which the solution set to SRM and HA can be derived from the joint-PCA are identified. In particular, if there exists a common template and a set of generalized rotation matrices through which datasets can be exactly aligned to the template, then for any number of features, SRM and HA solutions can be readily derived from the joint-PCA of datasets. Not surprisingly, this assumption fails to hold for complex multi-datasets, e.g., multi-subject fMRI datasets. We show that if the desired conditions are not satisfied, joint-PCA can easily over-fit to the training data when the dimension of the projected space is high (~> 50). We also examine how well low-dimensional matrix factorization can be computed using gradient descent-type algorithms using Google’s TensorFlow library.
{"title":"Multi-dataset Low-rank Matrix Factorization","authors":"Hossein Valavi, P. Ramadge","doi":"10.1109/CISS.2019.8692932","DOIUrl":"https://doi.org/10.1109/CISS.2019.8692932","url":null,"abstract":"Low-rank matrix factorization can reveal fundamental structure in data. For example, joint-PCA on multi-datasets can find a joint, lower-dimensional representation of the data. Recently other similar matrix factorization methods have been introduced for multi-dataset analysis, e.g., the shared response model (SRM) and hyperalignment (HA). We provide a comparison of these methods with joint-PCA that highlights similarities and differences. Necessary and sufficient conditions under which the solution set to SRM and HA can be derived from the joint-PCA are identified. In particular, if there exists a common template and a set of generalized rotation matrices through which datasets can be exactly aligned to the template, then for any number of features, SRM and HA solutions can be readily derived from the joint-PCA of datasets. Not surprisingly, this assumption fails to hold for complex multi-datasets, e.g., multi-subject fMRI datasets. We show that if the desired conditions are not satisfied, joint-PCA can easily over-fit to the training data when the dimension of the projected space is high (~> 50). We also examine how well low-dimensional matrix factorization can be computed using gradient descent-type algorithms using Google’s TensorFlow library.","PeriodicalId":123696,"journal":{"name":"2019 53rd Annual Conference on Information Sciences and Systems (CISS)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116246958","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：