2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS)最新文献

筛选
英文 中文
Using a map-based encoding to evolve plastic neural networks 使用基于地图的编码来进化塑性神经网络
2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS) Pub Date : 2011-04-11 DOI: 10.1109/EAIS.2011.5945909
Paul Tonelli, Jean-Baptiste Mouret
{"title":"Using a map-based encoding to evolve plastic neural networks","authors":"Paul Tonelli, Jean-Baptiste Mouret","doi":"10.1109/EAIS.2011.5945909","DOIUrl":"https://doi.org/10.1109/EAIS.2011.5945909","url":null,"abstract":"Many controllers for complex agents have been successfully generated by automatically desiging artificial neural networks with evolutionary algorithms. However, typical evolved neural networks are not able to adapt themselves online, making them unable to perform tasks that require online adaptation. Nature solved this problem on animals with plastic nervous systems. Inpired by neuroscience models of plastic neural-network, the present contribution proposes to use a combination of Hebbian learning, neuro-modulation and a a generative map-based encoding. We applied the proposed approach on a problem from operant conditioning (a Skinner box), in which numerous different association rules can be learned. Results show that the map-based encoding scaled up better than a classic direct encoding on this task. Evolving neural networks using a map-based generative encoding also lead to networks that works with most rule sets even when the evolution is done on a small subset of all the possible cases. Such a generative encoding therefore appears as a key to improve the generalization abilities of evolved adaptive neural networks.","PeriodicalId":243348,"journal":{"name":"2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"198200 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115187308","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}
引用次数: 11
Content-based audio classification using collective network of binary classifiers 基于内容的音频分类使用二元分类器的集体网络
2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS) Pub Date : 2011-04-11 DOI: 10.1109/EAIS.2011.5945911
Toni Mäkinen, S. Kiranyaz, M. Gabbouj
{"title":"Content-based audio classification using collective network of binary classifiers","authors":"Toni Mäkinen, S. Kiranyaz, M. Gabbouj","doi":"10.1109/EAIS.2011.5945911","DOIUrl":"https://doi.org/10.1109/EAIS.2011.5945911","url":null,"abstract":"In this paper, a novel collective network of binary classifiers (CNBC) framework is presented for content-based audio classification. The topic has been studied in several publications before, but in many cases the number of different classification categories is quite limited and needed to be fixed a priori. We focus our efforts to increase both the classification accuracy and the number of classes, as well as to create a scalable network design, which allows introducing new audio classes incrementally. The approach is based on dividing a major classification problem into several networks of binary classifiers (NBCs), where each NBC adapts its internal topology according to the classification problem at hand, by using evolutionary Artificial Neural Networks (ANNs). In the current work, feed-forward ANNs, or the so-called Multilayer Perceptrons (MLPs), are evolved within an architecture space, where a stochastic optimization is applied to seek for the optimal classifier configuration and parameters. The performance evaluations of the proposed framework over an 8-class benchmark audio database demonstrate its scalability and notable potential, as classification error rates of less than 9% are achieved.","PeriodicalId":243348,"journal":{"name":"2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128384916","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}
引用次数: 4
A genetic algorithm for self-optimization in safety-critical resource-flow systems 安全关键型资源流系统自优化的遗传算法
2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS) Pub Date : 2011-04-11 DOI: 10.1109/EAIS.2011.5945915
Florian Siefert, Florian Nafz, H. Seebach, W. Reif
{"title":"A genetic algorithm for self-optimization in safety-critical resource-flow systems","authors":"Florian Siefert, Florian Nafz, H. Seebach, W. Reif","doi":"10.1109/EAIS.2011.5945915","DOIUrl":"https://doi.org/10.1109/EAIS.2011.5945915","url":null,"abstract":"Organic Computing tries to tackle the rising complexity of systems by developing mechanisms and techniques that allow a system to self-organize and possess life-like behavior. The introduction of self-x properties also brings uncertainty and makes the systems unpredictable. Therefore, these systems are hardly used in safety-critical domains and their acceptance is low. If those systems should also profit from the benefits of self-x properties, behavioral guarantees must be provided. In this paper, a genetic algorithm for the self-optimization of resource-flow systems is presented. Further, its integration into an architecture which allows to provide behavioral guarantees is shown.","PeriodicalId":243348,"journal":{"name":"2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128917759","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}
引用次数: 3
Kernel evolution for support vector classification 支持向量分类的核演化
2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS) Pub Date : 2011-04-11 DOI: 10.1109/EAIS.2011.5945924
Mehrdad Alizadeh, M. Ebadzadeh
{"title":"Kernel evolution for support vector classification","authors":"Mehrdad Alizadeh, M. Ebadzadeh","doi":"10.1109/EAIS.2011.5945924","DOIUrl":"https://doi.org/10.1109/EAIS.2011.5945924","url":null,"abstract":"Support vector machines (SVMs) have been used in a variety of classification tasks. SVMs undoubtedly are one of the most effective classifiers in several data mining applications. Determination of a kernel function and related parameters has been a bottleneck for this group of classifiers. In this paper a novel approach is proposed to use genetic programming (GP) to design domain-specific and optimal kernel functions for support vector classification (SVC) which automatically adjusts the parameters. Complex low dimensional mapping function is evolved using GP to construct an optimal linear and Gaussian kernel functions in new feature space. By using the principled kernel closure properties, these basic kernels are then used to evolve more optimal kernels. To evaluate the proposed method, benchmark datasets from UCI are applied. The result indicates that for some cases the proposed methods can find a more optimal solution than evolving known kernels.","PeriodicalId":243348,"journal":{"name":"2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126369162","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}
引用次数: 3
Self-growing applications from abstract architectures an application to data-mediation systems 从抽象体系结构到数据中介系统的自增长应用程序
2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS) Pub Date : 2011-04-11 DOI: 10.1109/EAIS.2011.5945907
A. Diaconescu, P. Lalanda
{"title":"Self-growing applications from abstract architectures an application to data-mediation systems","authors":"A. Diaconescu, P. Lalanda","doi":"10.1109/EAIS.2011.5945907","DOIUrl":"https://doi.org/10.1109/EAIS.2011.5945907","url":null,"abstract":"Imagine a distributed mediation application consisting of hundreds of thousands of interconnected nodes, collecting data from millions of pervasive sensors, processing data and delivering it to a myriad of business services. This application takes the form of an acyclic, directed graph. Its shape must continually adapt in response to changes in sensor availability, network layout and business objectives. This involves dynamically adding, configuring, migrating and removing graph nodes. A centralised Observer/Controller, or Autonomic Manager (AM), that controls lifecycle operations for the entire graph would neither scale with the system's size and adaptation frequency, nor survive in unpredictable environments. This paper proposes a decentralised solution for enabling mediation applications to self-grow and to self-adapt their shapes and behaviours. In this approach, applications can autonomously grow into instances of a predefined, abstract architectural model and continually adapt to their execution conditions. A proof-of-concept prototype was developed using a Java-based, Service Oriented Component technology - iPOJO / OSGi. Experimental results from a Home Monitoring data-mediation scenario show the applicability and viability of our approach. We believe that the proposed framework will enable applications to autonomously grow and survive in volatile execution environments, over extended time periods.","PeriodicalId":243348,"journal":{"name":"2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"544 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133691996","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}
引用次数: 3
Evolution of an adaptive unsupervised speech controlled system 自适应无监督语音控制系统的演化
2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS) Pub Date : 2011-04-11 DOI: 10.1109/EAIS.2011.5945906
T. Herbig, F. Gerl, W. Minker
{"title":"Evolution of an adaptive unsupervised speech controlled system","authors":"T. Herbig, F. Gerl, W. Minker","doi":"10.1109/EAIS.2011.5945906","DOIUrl":"https://doi.org/10.1109/EAIS.2011.5945906","url":null,"abstract":"In this paper we present a self-learning speech controlled system comprising speech recognition, speaker identification and speaker adaptation. Our goal is the automatic personalization of speech controlled devices for some five recurring speakers without the requirement of a speaker specific training. We present a novel approach to detect unknown speakers based on a unified speech and speaker model. New users are detected in an unsupervised manner based on only a few utterances. New speaker profiles are initialized without any additional intervention of the user. Each profile is continuously adapted by tracking the speaker identity on successive utterances to enhance future speech recognition and speaker identification. Experiments on the evolution of such a system were carried out on a subset of the SPEECON database. The results show that in the long run the system produces adaptation profiles which give improvements in speech recognition rate only slightly lower than supervised or closed-set systems.","PeriodicalId":243348,"journal":{"name":"2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134230177","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}
引用次数: 3
A Simplified Structure Evolving Method for Fuzzy System structure learning 一种用于模糊系统结构学习的简化结构演化方法
2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS) Pub Date : 2011-04-11 DOI: 10.1109/EAIS.2011.5945914
Di Wang, Xiao-Jun Zeng, J. Keane
{"title":"A Simplified Structure Evolving Method for Fuzzy System structure learning","authors":"Di Wang, Xiao-Jun Zeng, J. Keane","doi":"10.1109/EAIS.2011.5945914","DOIUrl":"https://doi.org/10.1109/EAIS.2011.5945914","url":null,"abstract":"This paper proposes a Simplified Structure Evolving Method (SSEM) for Fuzzy Systems, which improves our previous work of Structure Evolving Learning Method for Fuzzy Systems (SELM [1]). SSEM keeps all the advantages of SELM [1] and improve SELM by starting with the simplest fuzzy rule set with only one fuzzy rule (instead of 2n fuzzy rules in SELM) as the starting point. By doing this SSEM is able to select the most efficient partitions and the most efficient attributes as well for system identification. This improvement enables fuzzy systems applicable to high dimensional problems. Benchmark examples with high dimension inputs are given to illustrate the advantages of the proposed algorithm.","PeriodicalId":243348,"journal":{"name":"2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"240 5","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120892891","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}
引用次数: 1
Kernel density estimation with stream data based on self-organizing map 基于自组织映射的流数据核密度估计
2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS) Pub Date : 2011-04-11 DOI: 10.1109/EAIS.2011.5945929
Haibo He, Yuan Cao
{"title":"Kernel density estimation with stream data based on self-organizing map","authors":"Haibo He, Yuan Cao","doi":"10.1109/EAIS.2011.5945929","DOIUrl":"https://doi.org/10.1109/EAIS.2011.5945929","url":null,"abstract":"We investigate the kernel density estimation (KDE) problem with stream data in this paper. Specifically, we analyze the characteristics of stream data density estimation, and propose an approach based on self-organizing map (SOM) to tackle the challenges of traditional KDE techniques for stream data analysis, such as computational cost, processing time, and memory requirement. Our proposed approach first generates SOMs for chunks of the data along the data streams, which obtains summaries of the data streams. Then, the probability density functions (pdfs) over arbitrary time periods along the data streams can be estimated with the generated SOMs. We compare our method with two other data stream KDE methods, the M-kernel and cluster kernel methods, in terms of accuracy and processing time. The simulation results illustrate the effectiveness and efficiency of the proposed algorithm.","PeriodicalId":243348,"journal":{"name":"2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114574413","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}
引用次数: 0
Evolving human activity classifier from sensor streams 从传感器流进化人类活动分类器
2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS) Pub Date : 2011-04-11 DOI: 10.1109/EAIS.2011.5945921
J. A. Iglesias, P. Angelov, Agapito Ledezma, A. Sanchis
{"title":"Evolving human activity classifier from sensor streams","authors":"J. A. Iglesias, P. Angelov, Agapito Ledezma, A. Sanchis","doi":"10.1109/EAIS.2011.5945921","DOIUrl":"https://doi.org/10.1109/EAIS.2011.5945921","url":null,"abstract":"Human activity recognition in intelligent environments is a very important task for many applications such as assisted living or surveillance. In order to make those environments sensitive to people, it is necessary to recognize and track the activities that they perform as part of their daily routines. Most of the current approaches for recognizing human activities do not consider the changes in how a human performs a specific activity. Those approaches rely on predefined activities which are represented as static models over time. In this paper, we propose an automated approach to track and recognize daily activities from sensor streams. Any activity is represented in this research as a sequence of raw sensors data. These sequences are treated using statistical methods in order to discover activity patterns. However, these patterns change due to the dynamic nature of human activities. Therefore, as the way to perform an activity is usually not fixed but it changes and evolves, we propose a human activity recognition method based on Evolving Systems.","PeriodicalId":243348,"journal":{"name":"2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114593572","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}
引用次数: 0
Collective network of evolutionary binary classifiers for content-based image retrieval 基于内容的图像检索的进化二元分类器集体网络
2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS) Pub Date : 2011-04-11 DOI: 10.1109/EAIS.2011.5945925
S. Kiranyaz, Stefan Uhlmann, Jenni Raitoharju, M. Gabbouj, T. Ince
{"title":"Collective network of evolutionary binary classifiers for content-based image retrieval","authors":"S. Kiranyaz, Stefan Uhlmann, Jenni Raitoharju, M. Gabbouj, T. Ince","doi":"10.1109/EAIS.2011.5945925","DOIUrl":"https://doi.org/10.1109/EAIS.2011.5945925","url":null,"abstract":"The content-based image retrieval (CBIR) has been an active research field for which several feature extraction, classification and retrieval techniques have been proposed up to date. However, when the database size grows larger, it is a common fact that the overall retrieval performance significantly deteriorates. In this paper, we propose collective network of (evolutionary) binary classifiers (CNBC) framework to achieve a high retrieval performance even though the training (ground truth) data may not be entirely present from the beginning and thus the system can only be trained incrementally. The CNBC framework basically adopts a “Divide and Conquer” type approach by allocating several networks of binary classifiers (NBCs) to discriminate each class and performs evolutionary search to find the optimal binary classifier (BC) in each NBC. In such an evolution session, the CNBC body can further dynamically adapt itself with each new incoming class/feature set without a full-scale re-training or re-configuration. Both visual and numerical performance evaluations of the proposed framework over benchmark image databases demonstrate its scalability; and a significant performance improvement is achieved over traditional retrieval techniques.","PeriodicalId":243348,"journal":{"name":"2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121702394","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}
引用次数: 4
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信