Pub Date : 2018-11-01Epub Date: 2019-02-21DOI: 10.1109/GlobalSIP.2018.8646507
Rui Xie, Fangyu Li, Zengyan Wang, WenZhan Song
The rapid convergence rate, high fidelity learning outcome and low computational cost are key targets in solving the learning problem of the complex physical system. Guided by physical laws of wave propagation, in full waveform inversion (FWI), we learn the subsurface images through optimizing the media velocity model in a large scale non-linear problem. In this paper, we combine randomized subsampling techniques with a second-order optimization algorithm to propose the Sub-Sampled Newton (SSN) method for learning velocity model of FWI. By incorporating the curvature information, SSN preserves comparable convergence rate to Newtons method and significantly reduces the iteration cost by approximating the Hessian matrix through a non-uniform subsampling scheme. The numerical experiments demonstrate that the proposed SSN method has a faster convergence rate, and achieves a more accurate velocity model in terms of mean squared error than commonly used methods.
{"title":"LARGE SCALE RANDOMIZED LEARNING GUIDED BY PHYSICAL LAWS WITH APPLICATIONS IN FULL WAVEFORM INVERSION.","authors":"Rui Xie, Fangyu Li, Zengyan Wang, WenZhan Song","doi":"10.1109/GlobalSIP.2018.8646507","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2018.8646507","url":null,"abstract":"<p><p>The rapid convergence rate, high fidelity learning outcome and low computational cost are key targets in solving the learning problem of the complex physical system. Guided by physical laws of wave propagation, in full waveform inversion (FWI), we learn the subsurface images through optimizing the media velocity model in a large scale non-linear problem. In this paper, we combine randomized subsampling techniques with a second-order optimization algorithm to propose the Sub-Sampled Newton (SSN) method for learning velocity model of FWI. By incorporating the curvature information, SSN preserves comparable convergence rate to Newtons method and significantly reduces the iteration cost by approximating the Hessian matrix through a non-uniform subsampling scheme. The numerical experiments demonstrate that the proposed SSN method has a faster convergence rate, and achieves a more accurate velocity model in terms of mean squared error than commonly used methods.</p>","PeriodicalId":91429,"journal":{"name":"... IEEE Global Conference on Signal and Information Processing. IEEE Global Conference on Signal and Information Processing","volume":"2018 ","pages":"66-70"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1109/GlobalSIP.2018.8646507","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"37338519","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 : 2017-11-01Epub Date: 2018-03-08DOI: 10.1109/GlobalSIP.2017.8308685
Shaobo Fang, Fengqing Zhu, Carol J Boushey, Edward J Delp
Measuring accurate dietary intake is considered to be an open research problem in the nutrition and health fields. Food portions estimation is a challenging problem as food preparation and consumption process pose large variations on food shapes and appearances. We use geometric model based technique to estimate food portions and further improve estimation accuracy using co-occurrence patterns. We estimate the food portion co-occurrence patterns from food images we collected from dietary studies using the mobile Food Record (mFR) system we developed. Co-occurrence patterns is used as prior knowledge to refine portion estimation results. We show that the portion estimation accuracy has been improved when incorporating the co-occurrence patterns as contextual information.
{"title":"THE USE OF CO-OCCURRENCE PATTERNS IN SINGLE IMAGE BASED FOOD PORTION ESTIMATION.","authors":"Shaobo Fang, Fengqing Zhu, Carol J Boushey, Edward J Delp","doi":"10.1109/GlobalSIP.2017.8308685","DOIUrl":"10.1109/GlobalSIP.2017.8308685","url":null,"abstract":"<p><p>Measuring accurate dietary intake is considered to be an open research problem in the nutrition and health fields. Food portions estimation is a challenging problem as food preparation and consumption process pose large variations on food shapes and appearances. We use geometric model based technique to estimate food portions and further improve estimation accuracy using co-occurrence patterns. We estimate the food portion co-occurrence patterns from food images we collected from dietary studies using the mobile Food Record (mFR) system we developed. Co-occurrence patterns is used as prior knowledge to refine portion estimation results. We show that the portion estimation accuracy has been improved when incorporating the co-occurrence patterns as contextual information.</p>","PeriodicalId":91429,"journal":{"name":"... IEEE Global Conference on Signal and Information Processing. IEEE Global Conference on Signal and Information Processing","volume":"2017 ","pages":"462-466"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6226047/pdf/nihms-995024.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"36665646","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 : 2015-01-01DOI: 10.1109/GlobalSIP.2015.7418423
F. Ali, S. Larbi
{"title":"Perceptual long-term harmonic plus noise modeling for speech data compression","authors":"F. Ali, S. Larbi","doi":"10.1109/GlobalSIP.2015.7418423","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2015.7418423","url":null,"abstract":"","PeriodicalId":91429,"journal":{"name":"... IEEE Global Conference on Signal and Information Processing. IEEE Global Conference on Signal and Information Processing","volume":"22 1","pages":"1372-1376"},"PeriodicalIF":0.0,"publicationDate":"2015-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77552204","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 : 2014-12-01Epub Date: 2015-02-09DOI: 10.1109/GlobalSIP.2014.7032351
Cheng Yang, Po-Yen Wu, John H Phan, May D Wang
RNA-seq data analysis pipelines are generally composed of sequence alignment, expression quantification, expression normalization, and differentially expressed gene (DEG) detection. Each step has numerous specific tools or algorithms, so we cannot explore all combinatorial pipelines and provide a comprehensive comparison of pipeline performance. To understand the mechanism of RNA-seq data analysis pipelines and provide some useful information for pipeline selection, we believe it is necessary to analyze the interactions among pipeline components. In this paper, by combining different alignment algorithms with the same quantification, normalization, and DEG detection tools, we construct nine RNA-seq pipelines to analyze the impact of RNA-seq alignment on downstream applications of gene expression estimates. Specifically, we find moderate linear correlation between the number of DEGs detected and the percentage of reads aligned with zero mismatch.
{"title":"The Impact of RNA-seq Alignment Pipeline on Detection of Differentially Expressed Genes.","authors":"Cheng Yang, Po-Yen Wu, John H Phan, May D Wang","doi":"10.1109/GlobalSIP.2014.7032351","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2014.7032351","url":null,"abstract":"RNA-seq data analysis pipelines are generally composed of sequence alignment, expression quantification, expression normalization, and differentially expressed gene (DEG) detection. Each step has numerous specific tools or algorithms, so we cannot explore all combinatorial pipelines and provide a comprehensive comparison of pipeline performance. To understand the mechanism of RNA-seq data analysis pipelines and provide some useful information for pipeline selection, we believe it is necessary to analyze the interactions among pipeline components. In this paper, by combining different alignment algorithms with the same quantification, normalization, and DEG detection tools, we construct nine RNA-seq pipelines to analyze the impact of RNA-seq alignment on downstream applications of gene expression estimates. Specifically, we find moderate linear correlation between the number of DEGs detected and the percentage of reads aligned with zero mismatch.","PeriodicalId":91429,"journal":{"name":"... IEEE Global Conference on Signal and Information Processing. IEEE Global Conference on Signal and Information Processing","volume":"2012 ","pages":"1376-1379"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1109/GlobalSIP.2014.7032351","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"34363350","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 : 2014-12-01DOI: 10.1109/GlobalSIP.2014.7032171
D. Kolossa
Human beings are highly effective at integrating multiple sources of uncertain information, and mounting evidence points to this integration being practically optimal in a Bayesian sense. Yet, in speech processing systems, the two central tasks of speech signal enhancement and of speech or phonetic-state recognition are often performed almost in isolation, with only estimates of mean values being exchanged between them. This paper describes concepts for enhancing the interface of these two systems, considering a range of appropriate probabilistic representations. Examples will illustrate how such interfaces can improve the quality of both components: On the one hand, more reliable pattern recognition can be attained, while on the other hand, enhanced signal quality is achieved when feeding back information from a pattern recognition stage to the signal preprocessing. This latter idea will be described using the example of twin-HMMs, audiovisual speech models that help to recover lost acoustic information by exploiting video data. Overall, it will be shown how broader, probabilistic interfaces between signal processing and pattern recognition can help to achieve better performance in real-world conditions, and to more closely approximate the Bayesian ideal of using all sources of information in accordance with their respective degree of reliability.
{"title":"Narrowing the gap: Probabilistic interfaces for signal enhancement and pattern recognition","authors":"D. Kolossa","doi":"10.1109/GlobalSIP.2014.7032171","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2014.7032171","url":null,"abstract":"Human beings are highly effective at integrating multiple sources of uncertain information, and mounting evidence points to this integration being practically optimal in a Bayesian sense. Yet, in speech processing systems, the two central tasks of speech signal enhancement and of speech or phonetic-state recognition are often performed almost in isolation, with only estimates of mean values being exchanged between them. This paper describes concepts for enhancing the interface of these two systems, considering a range of appropriate probabilistic representations. Examples will illustrate how such interfaces can improve the quality of both components: On the one hand, more reliable pattern recognition can be attained, while on the other hand, enhanced signal quality is achieved when feeding back information from a pattern recognition stage to the signal preprocessing. This latter idea will be described using the example of twin-HMMs, audiovisual speech models that help to recover lost acoustic information by exploiting video data. Overall, it will be shown how broader, probabilistic interfaces between signal processing and pattern recognition can help to achieve better performance in real-world conditions, and to more closely approximate the Bayesian ideal of using all sources of information in accordance with their respective degree of reliability.","PeriodicalId":91429,"journal":{"name":"... IEEE Global Conference on Signal and Information Processing. IEEE Global Conference on Signal and Information Processing","volume":"41 1","pages":"517-521"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74815437","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 : 2014-01-01DOI: 10.1109/GlobalSIP.2014.7032162
Shunqiao Sun, A. Petropulu
{"title":"On waveform design for MIMO radar with matrix completion","authors":"Shunqiao Sun, A. Petropulu","doi":"10.1109/GlobalSIP.2014.7032162","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2014.7032162","url":null,"abstract":"","PeriodicalId":91429,"journal":{"name":"... IEEE Global Conference on Signal and Information Processing. IEEE Global Conference on Signal and Information Processing","volume":"1 1","pages":"473-477"},"PeriodicalIF":0.0,"publicationDate":"2014-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88877905","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 : 2014-01-01DOI: 10.1109/GlobalSIP.2014.7032069
Zain-ul-Abdin, Mingkun Yang
{"title":"Dataflow programming of real-time radar signal processing on manycores","authors":"Zain-ul-Abdin, Mingkun Yang","doi":"10.1109/GlobalSIP.2014.7032069","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2014.7032069","url":null,"abstract":"","PeriodicalId":91429,"journal":{"name":"... IEEE Global Conference on Signal and Information Processing. IEEE Global Conference on Signal and Information Processing","volume":"2 1","pages":"15-19"},"PeriodicalIF":0.0,"publicationDate":"2014-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74406972","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 : 2013-12-01DOI: 10.1109/GlobalSIP.2013.6736791
C. Bouman, R. Nowak, A. Scaglione
Welcome to Austin, Texas for the inaugural IEEE Global Conference on Signal and Information Processing. GlobalSIP is a new flagship IEEE Signal Processing Society conference that targets hot topics and up-and-coming themes in signal and information processing. GlobalSIP is organized differently from other IEEE SPS meetings to encourage new SPS research directions and to foster emerging areas.
{"title":"Technical program overview","authors":"C. Bouman, R. Nowak, A. Scaglione","doi":"10.1109/GlobalSIP.2013.6736791","DOIUrl":"https://doi.org/10.1109/GlobalSIP.2013.6736791","url":null,"abstract":"Welcome to Austin, Texas for the inaugural IEEE Global Conference on Signal and Information Processing. GlobalSIP is a new flagship IEEE Signal Processing Society conference that targets hot topics and up-and-coming themes in signal and information processing. GlobalSIP is organized differently from other IEEE SPS meetings to encourage new SPS research directions and to foster emerging areas.","PeriodicalId":91429,"journal":{"name":"... IEEE Global Conference on Signal and Information Processing. IEEE Global Conference on Signal and Information Processing","volume":"31 1 1","pages":"6736791"},"PeriodicalIF":0.0,"publicationDate":"2013-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82513082","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