Pub Date : 2002-08-07DOI: 10.1109/IJCNN.2002.1007557
C. Diehl, J. Hampshire
To conduct real-time video surveillance using low-cost commercial off-the-shelf hardware, system designers typically define the classifiers prior to the deployment of the system so that the performance of the system can be optimized for a particular mission. This implies the system is restricted to interpreting activity in the environment in terms of the original context specified. Ideally the system should allow the user to provide additional context in an incremental fashion as conditions change. Given the volumes of data produced by the system, it is impractical for the user to periodically review and label a significant fraction of the available data. We explore a strategy for designing a real-time object classification process that aids the user in identifying novel, informative examples for efficient incremental learning.
{"title":"Real-time object classification and novelty detection for collaborative video surveillance","authors":"C. Diehl, J. Hampshire","doi":"10.1109/IJCNN.2002.1007557","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1007557","url":null,"abstract":"To conduct real-time video surveillance using low-cost commercial off-the-shelf hardware, system designers typically define the classifiers prior to the deployment of the system so that the performance of the system can be optimized for a particular mission. This implies the system is restricted to interpreting activity in the environment in terms of the original context specified. Ideally the system should allow the user to provide additional context in an incremental fashion as conditions change. Given the volumes of data produced by the system, it is impractical for the user to periodically review and label a significant fraction of the available data. We explore a strategy for designing a real-time object classification process that aids the user in identifying novel, informative examples for efficient incremental learning.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115038780","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1005448
C. H. Hall Barbosa, B. Melo, M. Vellasco, M. Pacheco, L. P. Vasconcellos
Different neural networks algorithms have already been employed on the inference of distillation column products quality. This paper applies Bayesian neural networks on the inference of diesel 85% ASTM distillation, and compares the results with traditional multilayer perceptrons. Also, several pre-processing and variables selection methods have been implemented and tested.
{"title":"Inference of distillation column products quality using Bayesian networks","authors":"C. H. Hall Barbosa, B. Melo, M. Vellasco, M. Pacheco, L. P. Vasconcellos","doi":"10.1109/IJCNN.2002.1005448","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1005448","url":null,"abstract":"Different neural networks algorithms have already been employed on the inference of distillation column products quality. This paper applies Bayesian neural networks on the inference of diesel 85% ASTM distillation, and compares the results with traditional multilayer perceptrons. Also, several pre-processing and variables selection methods have been implemented and tested.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"422 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116011964","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1007642
Xiang Cao, P. N. Suganthan
This paper describes a hierarchical overlapped architecture (HOGNG) based upon the growing neural gas (GNG) network. The proposed architecture combines the unsupervised and supervised learning schemes in GNG. This novel network model was used to perform automatic video shot detection and motion characterization. Experimental results are presented to show the good classification accuracy of the proposed algorithm on real MPEG video sequences.
{"title":"Hierarchical overlapped growing neural gas networks with applications to video shot detection and motion characterization","authors":"Xiang Cao, P. N. Suganthan","doi":"10.1109/IJCNN.2002.1007642","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1007642","url":null,"abstract":"This paper describes a hierarchical overlapped architecture (HOGNG) based upon the growing neural gas (GNG) network. The proposed architecture combines the unsupervised and supervised learning schemes in GNG. This novel network model was used to perform automatic video shot detection and motion characterization. Experimental results are presented to show the good classification accuracy of the proposed algorithm on real MPEG video sequences.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116308641","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1005556
P. Navarrete, Javier Ruiz-del-Solar
An interactive face retrieval system that uses self-organizing maps and user feedback is described. The system solves some problems of related content-based image retrieval systems: non-existence of trivial high-level human descriptions of the images and the gap between the high-level descriptions and the low-level features used to index the images.
{"title":"Interactive face retrieval using self-organizing maps","authors":"P. Navarrete, Javier Ruiz-del-Solar","doi":"10.1109/IJCNN.2002.1005556","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1005556","url":null,"abstract":"An interactive face retrieval system that uses self-organizing maps and user feedback is described. The system solves some problems of related content-based image retrieval systems: non-existence of trivial high-level human descriptions of the images and the gap between the high-level descriptions and the low-level features used to index the images.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"194 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116467181","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1007777
T. Draelos, D. Duggan, M. Collins, D. C. Wunsch
We explore adaptive critic designs for host-based intrusion detection because of their utilization of reinforcement learning, which allows learning exploits that are difficult to pinpoint in sensor data. Results on Solaris basic security module audit data demonstrate an ability to learn to distinguish between clean and exploit data.
{"title":"Adaptive critic designs for host-based intrusion detection","authors":"T. Draelos, D. Duggan, M. Collins, D. C. Wunsch","doi":"10.1109/IJCNN.2002.1007777","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1007777","url":null,"abstract":"We explore adaptive critic designs for host-based intrusion detection because of their utilization of reinforcement learning, which allows learning exploits that are difficult to pinpoint in sensor data. Results on Solaris basic security module audit data demonstrate an ability to learn to distinguish between clean and exploit data.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123495618","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1007489
R. Dara, S. C. Kremer, D. Stacey
We show the use of a self organizing map to cluster unlabeled data and to infer possible labelings from the clusters. Our inferred labels are presented to a multilayer perceptron along with labeled data, performance is improved over using only the labeled data. Results are presented for a number of popular real-world benchmark problems from domains other than text. This shows one way in which unlabeled data can be used to enhance supervised learning in a general-purpose neural network.
{"title":"Clustering unlabeled data with SOMs improves classification of labeled real-world data","authors":"R. Dara, S. C. Kremer, D. Stacey","doi":"10.1109/IJCNN.2002.1007489","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1007489","url":null,"abstract":"We show the use of a self organizing map to cluster unlabeled data and to infer possible labelings from the clusters. Our inferred labels are presented to a multilayer perceptron along with labeled data, performance is improved over using only the labeled data. Results are presented for a number of popular real-world benchmark problems from domains other than text. This shows one way in which unlabeled data can be used to enhance supervised learning in a general-purpose neural network.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"130 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123643024","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1007681
F. Tordini, F. Piazza
The possibility of introducing a-priori information into multichannel blind deconvolution algorithms is investigated. The maximum likelihood (ML) approach allows one to introduce an important feature of the voice, namely the pitch, naturally into the 'blind' model, removing the nonlinearity and showing the advantages of productive contaminations by such related research fields as computer-aided sound analysis (CASA) and Bayesian theory.
{"title":"A semi-blind approach to the separation of real world speech mixtures","authors":"F. Tordini, F. Piazza","doi":"10.1109/IJCNN.2002.1007681","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1007681","url":null,"abstract":"The possibility of introducing a-priori information into multichannel blind deconvolution algorithms is investigated. The maximum likelihood (ML) approach allows one to introduce an important feature of the voice, namely the pitch, naturally into the 'blind' model, removing the nonlinearity and showing the advantages of productive contaminations by such related research fields as computer-aided sound analysis (CASA) and Bayesian theory.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122080615","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1007545
Ji He, A. Tan, C. Tan
Proposes an ART-based neural architecture known as ART-C (ART under constraints) that performs online clustering of pattern sequences subject to the constraints on the recognition category representation. Experiments on two real-life data sets show that ART-C produces reasonably good clustering qualities, with the added advantage of high efficiency.
提出了一种基于ART的神经网络架构ART- c (ART under constraints)，在识别类别表示的约束下对模式序列进行在线聚类。在两个真实数据集上的实验表明，ART-C产生了相当好的聚类质量，并且具有高效率的优势。
{"title":"ART-C: a neural architecture for self-organization under constraints","authors":"Ji He, A. Tan, C. Tan","doi":"10.1109/IJCNN.2002.1007545","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1007545","url":null,"abstract":"Proposes an ART-based neural architecture known as ART-C (ART under constraints) that performs online clustering of pattern sequences subject to the constraints on the recognition category representation. Experiments on two real-life data sets show that ART-C produces reasonably good clustering qualities, with the added advantage of high efficiency.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120837348","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1007648
Konrad Paul Kording, Peter König, David Klein
It is largely unknown how the properties of the auditory system relate to the properties of natural sounds. Here, we analyze representations of simulated neurons that have optimally sparse activity in response to spectro-temporal speech data. These representations share important properties with the auditory neurons determined in electrophysiological experiments.
{"title":"Learning of sparse auditory receptive fields","authors":"Konrad Paul Kording, Peter König, David Klein","doi":"10.1109/IJCNN.2002.1007648","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1007648","url":null,"abstract":"It is largely unknown how the properties of the auditory system relate to the properties of natural sounds. Here, we analyze representations of simulated neurons that have optimally sparse activity in response to spectro-temporal speech data. These representations share important properties with the auditory neurons determined in electrophysiological experiments.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123934438","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1007663
E. H. Shiguemori, J.D. Simoes de Silva, H.F. Campos-Velho
This paper describes a neural network approach to the inverse problem of determining the initial temperature distribution on a slab with adiabatic boundary conditions, from transient temperature distribution, obtained at a given time. Two neural network architectures have been proposed to address the problem: the multilayer perceptron with backpropagation and radial basis functions (RBF), both trained with the whole temperature history mapping. The conducted simulations showed RBF networks present better solutions, faster training, but higher noise sensitiveness, as compared to the multilayer perceptron with backpropagation.
{"title":"Neural network systems for estimating the initial condition in a heat conduction problem","authors":"E. H. Shiguemori, J.D. Simoes de Silva, H.F. Campos-Velho","doi":"10.1109/IJCNN.2002.1007663","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1007663","url":null,"abstract":"This paper describes a neural network approach to the inverse problem of determining the initial temperature distribution on a slab with adiabatic boundary conditions, from transient temperature distribution, obtained at a given time. Two neural network architectures have been proposed to address the problem: the multilayer perceptron with backpropagation and radial basis functions (RBF), both trained with the whole temperature history mapping. The conducted simulations showed RBF networks present better solutions, faster training, but higher noise sensitiveness, as compared to the multilayer perceptron with backpropagation.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124668922","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：