Pub Date : 2022-01-01DOI: 10.1109/BigComp54360.2022.00012
Jorgi Luiz Bolonhezi Dias, Leandro Batista de Almeida, L. Albini
{"title":"Reducing Hadoop 3.x energy consumption through Energy Efficient Ethernet","authors":"Jorgi Luiz Bolonhezi Dias, Leandro Batista de Almeida, L. Albini","doi":"10.1109/BigComp54360.2022.00012","DOIUrl":"https://doi.org/10.1109/BigComp54360.2022.00012","url":null,"abstract":"","PeriodicalId":93400,"journal":{"name":"... International Conference on Big Data and Smart Computing. International Conference on Big Data and Smart Computing","volume":"20 1","pages":"9-14"},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84814066","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 : 2021-01-01Epub Date: 2021-03-10DOI: 10.1109/bigcomp51126.2021.00021
Tianyi Yao, Daniel LeJeune, Hamid Javadi, Richard G Baraniuk, Genevera I Allen
Ridge-like regularization often leads to improved generalization performance of machine learning models by mitigating overfitting. While ridge-regularized machine learning methods are widely used in many important applications, direct training via optimization could become challenging in huge data scenarios with millions of examples and features. We tackle such challenges by proposing a general approach that achieves ridge-like regularization through implicit techniques named Minipatch Ridge (MPRidge). Our approach is based on taking an ensemble of coefficients of unregularized learners trained on many tiny, random subsamples of both the examples and features of the training data, which we call minipatches. We empirically demonstrate that MPRidge induces an implicit ridge-like regularizing effect and performs nearly the same as explicit ridge regularization for a general class of predictors including logistic regression, SVM, and robust regression. Embarrassingly parallelizable, MPRidge provides a computationally appealing alternative to inducing ridge-like regularization for improving generalization performance in challenging big-data settings.
{"title":"Minipatch Learning as Implicit Ridge-Like Regularization.","authors":"Tianyi Yao, Daniel LeJeune, Hamid Javadi, Richard G Baraniuk, Genevera I Allen","doi":"10.1109/bigcomp51126.2021.00021","DOIUrl":"https://doi.org/10.1109/bigcomp51126.2021.00021","url":null,"abstract":"<p><p>Ridge-like regularization often leads to improved generalization performance of machine learning models by mitigating overfitting. While ridge-regularized machine learning methods are widely used in many important applications, direct training via optimization could become challenging in huge data scenarios with millions of examples and features. We tackle such challenges by proposing a general approach that achieves ridge-like regularization through implicit techniques named Minipatch Ridge (MPRidge). Our approach is based on taking an ensemble of coefficients of unregularized learners trained on many tiny, random subsamples of both the examples and features of the training data, which we call minipatches. We empirically demonstrate that MPRidge induces an implicit ridge-like regularizing effect and performs nearly the same as explicit ridge regularization for a general class of predictors including logistic regression, SVM, and robust regression. Embarrassingly parallelizable, MPRidge provides a computationally appealing alternative to inducing ridge-like regularization for improving generalization performance in challenging big-data settings.</p>","PeriodicalId":93400,"journal":{"name":"... International Conference on Big Data and Smart Computing. International Conference on Big Data and Smart Computing","volume":"2021 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1109/bigcomp51126.2021.00021","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"39865226","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-01-01Epub Date: 2021-03-10DOI: 10.1109/bigcomp51126.2021.00023
Mohammad Taha Toghani, Genevera I Allen
Boosting methods are among the best general-purpose and off-the-shelf machine learning approaches, gaining widespread popularity. In this paper, we seek to develop a boosting method that yields comparable accuracy to popular AdaBoost and gradient boosting methods, yet is faster computationally and whose solution is more interpretable. We achieve this by developing MP-Boost, an algorithm loosely based on AdaBoost that learns by adaptively selecting small subsets of instances and features, or what we term minipatches (MP), at each iteration. By sequentially learning on tiny subsets of the data, our approach is computationally faster than other classic boosting algorithms. Also as it progresses, MP-Boost adaptively learns a probability distribution on the features and instances that upweight the most important features and challenging instances, hence adaptively selecting the most relevant minipatches for learning. These learned probability distributions also aid in interpretation of our method. We empirically demonstrate the interpretability, comparative accuracy, and computational time of our approach on a variety of binary classification tasks.
{"title":"MP-Boost: Minipatch Boosting via Adaptive Feature and Observation Sampling.","authors":"Mohammad Taha Toghani, Genevera I Allen","doi":"10.1109/bigcomp51126.2021.00023","DOIUrl":"https://doi.org/10.1109/bigcomp51126.2021.00023","url":null,"abstract":"<p><p>Boosting methods are among the best general-purpose and off-the-shelf machine learning approaches, gaining widespread popularity. In this paper, we seek to develop a boosting method that yields comparable accuracy to popular AdaBoost and gradient boosting methods, yet is faster computationally and whose solution is more interpretable. We achieve this by developing MP-Boost, an algorithm loosely based on AdaBoost that learns by adaptively selecting small subsets of instances and features, or what we term <i>minipatches</i> (MP), at each iteration. By sequentially learning on tiny subsets of the data, our approach is computationally faster than other classic boosting algorithms. Also as it progresses, MP-Boost adaptively learns a probability distribution on the features and instances that upweight the most important features and challenging instances, hence adaptively selecting the most relevant minipatches for learning. These learned probability distributions also aid in interpretation of our method. We empirically demonstrate the interpretability, comparative accuracy, and computational time of our approach on a variety of binary classification tasks.</p>","PeriodicalId":93400,"journal":{"name":"... International Conference on Big Data and Smart Computing. International Conference on Big Data and Smart Computing","volume":"2021 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1109/bigcomp51126.2021.00023","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"39677761","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-01-01DOI: 10.1109/BigComp51126.2021.00064
N. Dinh, Gon-Woo Kim
{"title":"Solid-State LiDAR based-SLAM: A Concise Review and Application","authors":"N. Dinh, Gon-Woo Kim","doi":"10.1109/BigComp51126.2021.00064","DOIUrl":"https://doi.org/10.1109/BigComp51126.2021.00064","url":null,"abstract":"","PeriodicalId":93400,"journal":{"name":"... International Conference on Big Data and Smart Computing. International Conference on Big Data and Smart Computing","volume":"59 1","pages":"302-305"},"PeriodicalIF":0.0,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82302084","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 : 2019-12-31DOI: 10.1515/9783110618372-016
{"title":"Anhang E: Anwendungsbeispiel 1","authors":"","doi":"10.1515/9783110618372-016","DOIUrl":"https://doi.org/10.1515/9783110618372-016","url":null,"abstract":"","PeriodicalId":93400,"journal":{"name":"... International Conference on Big Data and Smart Computing. International Conference on Big Data and Smart Computing","volume":"23 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80047923","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 : 2019-12-31DOI: 10.1515/9783110618372-002
Numărul, Secţia, Construcţii DE Maşini
Strict pollution norms make car’s manufacturers to promote electrical and hybrid vehicles (EVs and HEVs). Recuperation system that can capture about 8-25% of the energy is represented by the regenerative brakes, it converts losses of kinetic energy of the car during braking in form of heat to electrical or chemical energy. Regenerative brakes can provide a good increase in car’s autonomy, especially in city driving cycles with high frequency of braking. Also, they reduce friction brakes wear and offer more precisely control of wheel braking torque. A survey in this paper presents basics, uses, types and braking strategies for existing regenerative braking systems with the purpose to identify and detail their limitations. Identifying key factors that have influence on system efficiency allows to elaborate optimizing techniques, strategies and algorithms, applied to existing systems in order to develop future concepts that will overcome current limitations.
{"title":"Abstract","authors":"Numărul, Secţia, Construcţii DE Maşini","doi":"10.1515/9783110618372-002","DOIUrl":"https://doi.org/10.1515/9783110618372-002","url":null,"abstract":"Strict pollution norms make car’s manufacturers to promote electrical and hybrid vehicles (EVs and HEVs). Recuperation system that can capture about 8-25% of the energy is represented by the regenerative brakes, it converts losses of kinetic energy of the car during braking in form of heat to electrical or chemical energy. Regenerative brakes can provide a good increase in car’s autonomy, especially in city driving cycles with high frequency of braking. Also, they reduce friction brakes wear and offer more precisely control of wheel braking torque. A survey in this paper presents basics, uses, types and braking strategies for existing regenerative braking systems with the purpose to identify and detail their limitations. Identifying key factors that have influence on system efficiency allows to elaborate optimizing techniques, strategies and algorithms, applied to existing systems in order to develop future concepts that will overcome current limitations.","PeriodicalId":93400,"journal":{"name":"... International Conference on Big Data and Smart Computing. International Conference on Big Data and Smart Computing","volume":"22 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91169026","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 : 2019-12-31DOI: 10.1515/9783110618372-008
{"title":"6. Anwendungsbeispiele","authors":"","doi":"10.1515/9783110618372-008","DOIUrl":"https://doi.org/10.1515/9783110618372-008","url":null,"abstract":"","PeriodicalId":93400,"journal":{"name":"... International Conference on Big Data and Smart Computing. International Conference on Big Data and Smart Computing","volume":"28 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82554737","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 : 2019-12-31DOI: 10.1515/9783110618372-fm
{"title":"Frontmatter","authors":"","doi":"10.1515/9783110618372-fm","DOIUrl":"https://doi.org/10.1515/9783110618372-fm","url":null,"abstract":"","PeriodicalId":93400,"journal":{"name":"... International Conference on Big Data and Smart Computing. International Conference on Big Data and Smart Computing","volume":"327 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86776778","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 : 2019-12-31DOI: 10.1515/9783110618372-007
{"title":"5. Vorgehen der praktischen Anwendung","authors":"","doi":"10.1515/9783110618372-007","DOIUrl":"https://doi.org/10.1515/9783110618372-007","url":null,"abstract":"","PeriodicalId":93400,"journal":{"name":"... International Conference on Big Data and Smart Computing. International Conference on Big Data and Smart Computing","volume":"112 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85841031","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 : 2019-12-31DOI: 10.1515/9783110618372-014
{"title":"Anhang C: Katalog der Anwendungsziele","authors":"","doi":"10.1515/9783110618372-014","DOIUrl":"https://doi.org/10.1515/9783110618372-014","url":null,"abstract":"","PeriodicalId":93400,"journal":{"name":"... International Conference on Big Data and Smart Computing. International Conference on Big Data and Smart Computing","volume":"76 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85251328","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号