Pub Date : 2019-02-01DOI: 10.1109/icaiic.2019.8669058
{"title":"ICAIIC 2019 Program Book","authors":"","doi":"10.1109/icaiic.2019.8669058","DOIUrl":"https://doi.org/10.1109/icaiic.2019.8669058","url":null,"abstract":"","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131184932","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 : 2019-02-01DOI: 10.1109/ICAIIC.2019.8669008
H. Pasandi, T. Nadeem
To cope with the emergence of new technologies, various device characteristics and application requirements, complex and custom design of high performance networking protocols is much needed. Networking protocols, practically, are designed through long-time and hard-work human efforts. However, these designed protocols, typically, have limited flexibility that results in non-optimal performance under several network scenarios and conditions. Therefore, replacing this inefficient human based designing process by a novel paradigm that enables rapid design of efficient, flexible and high performance protocols that intelligently adapt to different device characteristics, application requirements, user objectives, and network conditions is highly desired. In this paper, we motivate the importance of a shift from human-driven protocol design process to a machine-based design. We propose a novel, self-managing and self-adaptive framework for automating MAC protocol design. As an example of such a framework, We design, implement, and evaluate AlphaMAC framework that learns to automate the design of efficient simple MAC protocols. We decouple MAC into a set of building blocks, and we are interested to see what blocks are selected by AlphaMAC in different scenarios, and whether the designed protocol is efficient. Our results show that AlphaMAC is able to select the efficient set of building blocks from ALOHA protocol building block set such that the designed protocol outperforms conventional ALOHA. We also discuss some of the challenges and limitations of realizing such a framework. We believe that the impact of the automated design of networking protocols on the network research and industrial community, and on developing networking services and applications would be significant.
{"title":"Challenges and Limitations in Automating the Design of MAC Protocols Using Machine-Learning","authors":"H. Pasandi, T. Nadeem","doi":"10.1109/ICAIIC.2019.8669008","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8669008","url":null,"abstract":"To cope with the emergence of new technologies, various device characteristics and application requirements, complex and custom design of high performance networking protocols is much needed. Networking protocols, practically, are designed through long-time and hard-work human efforts. However, these designed protocols, typically, have limited flexibility that results in non-optimal performance under several network scenarios and conditions. Therefore, replacing this inefficient human based designing process by a novel paradigm that enables rapid design of efficient, flexible and high performance protocols that intelligently adapt to different device characteristics, application requirements, user objectives, and network conditions is highly desired. In this paper, we motivate the importance of a shift from human-driven protocol design process to a machine-based design. We propose a novel, self-managing and self-adaptive framework for automating MAC protocol design. As an example of such a framework, We design, implement, and evaluate AlphaMAC framework that learns to automate the design of efficient simple MAC protocols. We decouple MAC into a set of building blocks, and we are interested to see what blocks are selected by AlphaMAC in different scenarios, and whether the designed protocol is efficient. Our results show that AlphaMAC is able to select the efficient set of building blocks from ALOHA protocol building block set such that the designed protocol outperforms conventional ALOHA. We also discuss some of the challenges and limitations of realizing such a framework. We believe that the impact of the automated design of networking protocols on the network research and industrial community, and on developing networking services and applications would be significant.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115025134","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 : 2019-02-01DOI: 10.1109/ICAIIC.2019.8669033
Aran V. Samson, A. Coronel
Musicology is a growing focus in computer science. Past research has had success in automatically generating music through learning-based agents [1] that make use of neural networks and through model and rule-based approaches [2]. These methods require a significant amount of information, either in the form of a large dataset for learning or a comprehensive set of rules based on musical concepts. This paper explores a model in which a minimal amount of musical information is needed to compose a desired style of music. This paper makes use of objectness, a concept directly derived from imagery and pattern recognition to extract specific musical objects from a single musical piece. This is then used as the foundation to produce a new generated musical piece that is similar in style to the original. The overall musical piece is generated through a partial evolution. This method eliminates the need for a large amount of pre-provided data and directly composes music based on a singular source piece.
{"title":"Reproducing Musicality: Detecting Musical Objects and Emulating Musicality Through Partial Evolution","authors":"Aran V. Samson, A. Coronel","doi":"10.1109/ICAIIC.2019.8669033","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8669033","url":null,"abstract":"Musicology is a growing focus in computer science. Past research has had success in automatically generating music through learning-based agents [1] that make use of neural networks and through model and rule-based approaches [2]. These methods require a significant amount of information, either in the form of a large dataset for learning or a comprehensive set of rules based on musical concepts. This paper explores a model in which a minimal amount of musical information is needed to compose a desired style of music. This paper makes use of objectness, a concept directly derived from imagery and pattern recognition to extract specific musical objects from a single musical piece. This is then used as the foundation to produce a new generated musical piece that is similar in style to the original. The overall musical piece is generated through a partial evolution. This method eliminates the need for a large amount of pre-provided data and directly composes music based on a singular source piece.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126873894","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 : 2019-02-01DOI: 10.1109/ICAIIC.2019.8669021
Jaehan Joo, Sinjin Jeong, Heetae Jin, Uhyeon Lee, Ji Young Yoon, S. Kim
In this paper, we propose a classification method of periodontal disease based on CNN. The data to used were the actual periodontal images and non-periodontal images. Data processing techniques such as resize, crop and zero-centralizing are used to improve data learning efficiency. The CNN Structure proposed in this paper has 224 × 224 × 3 size image as input data and 4 outputs according to periodontal state. We also use momentum optimization technique for neural network optimization.
{"title":"Periodontal Disease Detection Using Convolutional Neural Networks","authors":"Jaehan Joo, Sinjin Jeong, Heetae Jin, Uhyeon Lee, Ji Young Yoon, S. Kim","doi":"10.1109/ICAIIC.2019.8669021","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8669021","url":null,"abstract":"In this paper, we propose a classification method of periodontal disease based on CNN. The data to used were the actual periodontal images and non-periodontal images. Data processing techniques such as resize, crop and zero-centralizing are used to improve data learning efficiency. The CNN Structure proposed in this paper has 224 × 224 × 3 size image as input data and 4 outputs according to periodontal state. We also use momentum optimization technique for neural network optimization.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121887146","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 : 2019-02-01DOI: 10.1109/ICAIIC.2019.8669003
Seohyun Lee, Hyuno Kim, M. Ishikawa, H. Higuchi
Tracking of intracellular organelle movement such as vesicle includes crucial information in biomedicine. To achieve more accurate three-dimensional localization of the target organelle, superresolution imaging microscopy and image processing methods have been developed and applied to many nanoscale tracking systems. Although such recent advances in microscopy imaging have enabled us to gather a tremendous amount of tracking data, the details of the movement including the interaction between cytoskeletons are not yet fully explained. In the present work, we suggest a machine learning approach to clarify the problem in tracking data analysis, as an initial trial to exploit artificial intelligence in distinguishing and classifying the detail features of the vesicle-cytoskeleton interactions.
{"title":"3D Nanoscale Tracking Data Analysis for Intracellular Organelle Movement using Machine Learning Approach","authors":"Seohyun Lee, Hyuno Kim, M. Ishikawa, H. Higuchi","doi":"10.1109/ICAIIC.2019.8669003","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8669003","url":null,"abstract":"Tracking of intracellular organelle movement such as vesicle includes crucial information in biomedicine. To achieve more accurate three-dimensional localization of the target organelle, superresolution imaging microscopy and image processing methods have been developed and applied to many nanoscale tracking systems. Although such recent advances in microscopy imaging have enabled us to gather a tremendous amount of tracking data, the details of the movement including the interaction between cytoskeletons are not yet fully explained. In the present work, we suggest a machine learning approach to clarify the problem in tracking data analysis, as an initial trial to exploit artificial intelligence in distinguishing and classifying the detail features of the vesicle-cytoskeleton interactions.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127735559","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}
The interference caused by the D2D users to conventional cellular users and vice versa are the main drawbacks of underlay D2D communications. In order to mitigate the interference issue, we propose a resource management scheme based on frequency reuse method. In this paper, we formulate the problem of sum throughput optimization problem where each multicast D2D group can reuse only one uplink cellular link at a time. Also, the fractional frequency reuse (FFR) technique is considered and assumed that the cellular users and D2D users can share the resources in a non-orthogonal fashion. The performance of the proposed scheme is evaluated through extensive simulations using Monte-Carlo simulation. Simulation results demonstrate that the proposed scheme outperforms the random resource management scheme without cell sectorization method.
{"title":"Interference Mitigation for Multicast D2D Communications Underlay Cellular Networks","authors":"Devarani Devi Ningombam, Chung-Ghiu Lee, Seokjoo Shin","doi":"10.1109/ICAIIC.2019.8668982","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8668982","url":null,"abstract":"The interference caused by the D2D users to conventional cellular users and vice versa are the main drawbacks of underlay D2D communications. In order to mitigate the interference issue, we propose a resource management scheme based on frequency reuse method. In this paper, we formulate the problem of sum throughput optimization problem where each multicast D2D group can reuse only one uplink cellular link at a time. Also, the fractional frequency reuse (FFR) technique is considered and assumed that the cellular users and D2D users can share the resources in a non-orthogonal fashion. The performance of the proposed scheme is evaluated through extensive simulations using Monte-Carlo simulation. Simulation results demonstrate that the proposed scheme outperforms the random resource management scheme without cell sectorization method.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121254858","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 : 2019-02-01DOI: 10.1109/ICAIIC.2019.8669024
Zih-Siang Lin, H. Hung, Hoang-Yang Lu
In this paper, a two-stage spectrum sensing scheme based on angle-of-arrival (AoA) estimation of multiple primary users is proposed. In the proposed scheme, the delay-and-sum beamformer is adopted at the first stage to determine the presence of the possibly existing PUs and their approximate AoA information. At the second stage, from the AoA of each being determined PU, the MUSIC spectrum is scanned in a small region around the estimated AoA to obtain accurate AoA estimates of possibly more nearby PUs that could be hidden at the first stage. Simulation results and complexity analysis show that the proposed method has better performance without increasing computation complexity, as compared to the central-symmetry-based feature detection method.
{"title":"A Two-Stage Spatial Spectrum Sensing Scheme Based on Angle of Arrival Estimation","authors":"Zih-Siang Lin, H. Hung, Hoang-Yang Lu","doi":"10.1109/ICAIIC.2019.8669024","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8669024","url":null,"abstract":"In this paper, a two-stage spectrum sensing scheme based on angle-of-arrival (AoA) estimation of multiple primary users is proposed. In the proposed scheme, the delay-and-sum beamformer is adopted at the first stage to determine the presence of the possibly existing PUs and their approximate AoA information. At the second stage, from the AoA of each being determined PU, the MUSIC spectrum is scanned in a small region around the estimated AoA to obtain accurate AoA estimates of possibly more nearby PUs that could be hidden at the first stage. Simulation results and complexity analysis show that the proposed method has better performance without increasing computation complexity, as compared to the central-symmetry-based feature detection method.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128748186","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 : 2019-02-01DOI: 10.1109/ICAIIC.2019.8668980
T. Polevaya, R. Ravodin, A. Filchenkov
Automatic diagnostics of skin lesions is an area of high interest. Identification of primary morphology in skin lesions could be a first step of an automatic diagnostics tool. We propose an end-to-end deep learning solution to the problem of classifying primary morphology images of types macule, nodule, papule and plaque. Experimental results show 0.775 accuracy on 4 classes and 0.8167 accuracy on 3 classes.
{"title":"Skin Lesion Primary Morphology Classification With End-To-End Deep Learning Network","authors":"T. Polevaya, R. Ravodin, A. Filchenkov","doi":"10.1109/ICAIIC.2019.8668980","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8668980","url":null,"abstract":"Automatic diagnostics of skin lesions is an area of high interest. Identification of primary morphology in skin lesions could be a first step of an automatic diagnostics tool. We propose an end-to-end deep learning solution to the problem of classifying primary morphology images of types macule, nodule, papule and plaque. Experimental results show 0.775 accuracy on 4 classes and 0.8167 accuracy on 3 classes.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124409050","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 : 2019-02-01DOI: 10.1109/ICAIIC.2019.8668987
Dong Huh, Taekyung Kim, Jaeil Kim
Object localization in video is to predict the location and image boundaries of objects of interest in sequential scenes. Despite numerous methods being developed for the task, there are still challenging issues, such as labor-intensive data preparation. In this paper, we propose a patch-wise approach with weak supervision to resolve those issues in the object localization. We first train an patch-wise object classifier based on convolutional neural network with simple labeling about object classes, instead of the bounding box annotation. Then, the object regions are estimated using the class activation maps of the classifier for each patch. The patch-wise classifier can learn more relevant features of objects from the patches containing various parts of them. In addition, background patches for weakly-supervised learning can be easily prepared. Experiments using the visual object tracking challenge data set showed that the patch-wise weakly supervised approach is effective in the object localization in video.
{"title":"Patch-wise Weakly Supervised Learning for Object Localization in Video","authors":"Dong Huh, Taekyung Kim, Jaeil Kim","doi":"10.1109/ICAIIC.2019.8668987","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8668987","url":null,"abstract":"Object localization in video is to predict the location and image boundaries of objects of interest in sequential scenes. Despite numerous methods being developed for the task, there are still challenging issues, such as labor-intensive data preparation. In this paper, we propose a patch-wise approach with weak supervision to resolve those issues in the object localization. We first train an patch-wise object classifier based on convolutional neural network with simple labeling about object classes, instead of the bounding box annotation. Then, the object regions are estimated using the class activation maps of the classifier for each patch. The patch-wise classifier can learn more relevant features of objects from the patches containing various parts of them. In addition, background patches for weakly-supervised learning can be easily prepared. Experiments using the visual object tracking challenge data set showed that the patch-wise weakly supervised approach is effective in the object localization in video.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116957802","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 : 2019-02-01DOI: 10.1109/ICAIIC.2019.8669026
Hang Liu, Xu Zhu, T. Fujii
Spectrum sensing of orthogonal frequency division multiplex (OFDM) system has always been a challenge in cognitive radios (CR), especially for users which utilize the full-duplex(FD) mode. In this paper, we propose an advanced FD spectrum sensing scheme which can be successfully performed even when encountering severely self-interference from the user terminal. On the basis of ”classification converted sensing” framework, the cyclostationary periodogram generated by OFDM pilots is deduced in the form of images. These images are then plugged into the convolutional neural networks (CNNs) for classifications due to CNN’s strength in image recognition. More importantly, to achieve spectrum sensing against the residual self-interference, as well as the noise pollution and channel fading, we use the adversarial training where a CR-specific, modified training database is proposed. In addition, we propose a design plan of the signal structure for the CR terminal transmitting, which can fit in the proposed spectrum sensing scheme while benefiting its own transmission. Simulation results proved our method possesses an excellent sensing capability for the full-duplex system while achieving higher detection accuracy over the conventional method.
{"title":"Cyclostationary based full-duplex spectrum sensing using adversarial training for convolutional neural networks","authors":"Hang Liu, Xu Zhu, T. Fujii","doi":"10.1109/ICAIIC.2019.8669026","DOIUrl":"https://doi.org/10.1109/ICAIIC.2019.8669026","url":null,"abstract":"Spectrum sensing of orthogonal frequency division multiplex (OFDM) system has always been a challenge in cognitive radios (CR), especially for users which utilize the full-duplex(FD) mode. In this paper, we propose an advanced FD spectrum sensing scheme which can be successfully performed even when encountering severely self-interference from the user terminal. On the basis of ”classification converted sensing” framework, the cyclostationary periodogram generated by OFDM pilots is deduced in the form of images. These images are then plugged into the convolutional neural networks (CNNs) for classifications due to CNN’s strength in image recognition. More importantly, to achieve spectrum sensing against the residual self-interference, as well as the noise pollution and channel fading, we use the adversarial training where a CR-specific, modified training database is proposed. In addition, we propose a design plan of the signal structure for the CR terminal transmitting, which can fit in the proposed spectrum sensing scheme while benefiting its own transmission. Simulation results proved our method possesses an excellent sensing capability for the full-duplex system while achieving higher detection accuracy over the conventional method.","PeriodicalId":273383,"journal":{"name":"2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115376396","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：