Pub Date : 2011-10-03DOI: 10.1109/IJCNN.2011.6033248
K. Ang, Z. Chin, Haihong Zhang, Cuntai Guan
The Filter Bank Common Spatial Pattern (FBCSP) algorithm employs multiple spatial filters to automatically select key temporal-spatial discriminative EEG characteristics and the Naïve Bayesian Parzen Window (NBPW) classifier using offline learning in EEG-based Brain-Computer Interfaces (BCI). However, it has yet to address the non-stationarity inherent in the EEG between the initial calibration session and subsequent online sessions. This paper presents the FBCSP that employs the NBPW classifier using online adaptive learning that augments the training data with available labeled data during online sessions. However, employing semi-supervised learning that simply augments the training data with available data using predicted labels can be detrimental to the classification accuracy. Hence, this paper presents the FBCSP using online semi-supervised learning that augments the training data with available data that matches the probabilistic model captured by the NBPW classifier using predicted labels. The performances of FBCSP using online adaptive and semi-supervised learning are evaluated on the BCI Competition IV datasets IIa and IIb and compared to the FBCSP using offline learning. The results showed that the FBCSP using online semi-supervised learning yielded relatively better session-to-session classification results compared against the FBCSP using offline learning. The FBCSP using online adaptive learning on true labels yielded the best results in both datasets, but the FBCSP using online semi-supervised learning on predicted labels is more practical in BCI applications where the true labels are not available.
滤波器组公共空间模式(Filter Bank Common Spatial Pattern, FBCSP)算法采用多个空间滤波器自动选择关键的时空判别性脑电特征，并在脑机接口(BCI)中采用离线学习的Naïve贝叶斯帕森窗(NBPW)分类器。然而，它还没有解决在初始校准会话和随后的在线会话之间的脑电图固有的非平稳性。本文提出了使用在线自适应学习的NBPW分类器的FBCSP，该分类器在在线会话期间使用可用的标记数据来增强训练数据。然而，使用半监督学习，即使用预测标签简单地用可用数据增强训练数据，可能会损害分类精度。因此，本文提出了使用在线半监督学习的FBCSP，该方法使用与NBPW分类器使用预测标签捕获的概率模型相匹配的可用数据来增强训练数据。在BCI Competition IV数据集IIa和IIb上评估了使用在线自适应和半监督学习的FBCSP的性能，并与使用离线学习的FBCSP进行了比较。结果表明，与使用离线学习的FBCSP相比，使用在线半监督学习的FBCSP产生了相对更好的会话到会话分类结果。在真实标签上使用在线自适应学习的FBCSP在两个数据集中都产生了最好的结果，但是在真实标签不可用的BCI应用中，在预测标签上使用在线半监督学习的FBCSP更实用。
{"title":"Filter Bank Common Spatial Pattern (FBCSP) algorithm using online adaptive and semi-supervised learning","authors":"K. Ang, Z. Chin, Haihong Zhang, Cuntai Guan","doi":"10.1109/IJCNN.2011.6033248","DOIUrl":"https://doi.org/10.1109/IJCNN.2011.6033248","url":null,"abstract":"The Filter Bank Common Spatial Pattern (FBCSP) algorithm employs multiple spatial filters to automatically select key temporal-spatial discriminative EEG characteristics and the Naïve Bayesian Parzen Window (NBPW) classifier using offline learning in EEG-based Brain-Computer Interfaces (BCI). However, it has yet to address the non-stationarity inherent in the EEG between the initial calibration session and subsequent online sessions. This paper presents the FBCSP that employs the NBPW classifier using online adaptive learning that augments the training data with available labeled data during online sessions. However, employing semi-supervised learning that simply augments the training data with available data using predicted labels can be detrimental to the classification accuracy. Hence, this paper presents the FBCSP using online semi-supervised learning that augments the training data with available data that matches the probabilistic model captured by the NBPW classifier using predicted labels. The performances of FBCSP using online adaptive and semi-supervised learning are evaluated on the BCI Competition IV datasets IIa and IIb and compared to the FBCSP using offline learning. The results showed that the FBCSP using online semi-supervised learning yielded relatively better session-to-session classification results compared against the FBCSP using offline learning. The FBCSP using online adaptive learning on true labels yielded the best results in both datasets, but the FBCSP using online semi-supervised learning on predicted labels is more practical in BCI applications where the true labels are not available.","PeriodicalId":415833,"journal":{"name":"The 2011 International Joint Conference on Neural Networks","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132311887","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 : 2011-10-03DOI: 10.1109/IJCNN.2011.6033478
S. Okada, T. Nishida
In this paper, we present a novel incremental clustering algorithm which assigns of a set of observations into clusters and learns the distance metric iteratively in an incremental manner. The proposed algorithm SOINN-AML is composed based on the Self-organizing Incremental Neural Network (Shen et al 2006), which represents the distribution of unlabeled data and reports a reasonable number of clusters. SOINN adopts a competitive Hebbian rule for each input signal, and distance between nodes is measured using the Euclidean distance. Such algorithms rely on the distance metric for the input data patterns. Distance Metric Learning (DML) learns a distance metric for the high dimensional input space of data that preserves the distance relation among the training data. DML is not performed for input space of data in SOINN based approaches. SOINN-AML learns input space of data by using the Adaptive Distance Metric Learning (AML) algorithm which is one of the DML algorithms. It improves the incremental clustering performance of the SOINN algorithm by optimizing the distance metric in the case that input data space is high dimensional. In experimental results, we evaluate the performance by using two artificial datasets, seven real datasets from the UCI dataset and three real image datasets. We have found that the proposed algorithm outperforms conventional algorithms including SOINN (Shen et al 2006) and Enhanced SOINN (Shen et al 2007). The improvement of clustering accuracy (NMI) is between 0.03 and 0.13 compared to state of the art SOINN based approaches.
本文提出了一种新的增量聚类算法，该算法将一组观测值分配到聚类中，并以增量的方式迭代学习距离度量。本文提出的SOINN-AML算法是基于自组织增量神经网络(Self-organizing Incremental Neural Network, Shen et al . 2006)组成的，它代表了未标记数据的分布，并报告了合理数量的聚类。SOINN对每个输入信号采用竞争Hebbian规则，节点间距离采用欧氏距离测量。这种算法依赖于输入数据模式的距离度量。距离度量学习(Distance Metric Learning, DML)是对数据的高维输入空间学习一种保持训练数据之间距离关系的距离度量。在基于SOINN的方法中，对数据的输入空间不执行DML。SOINN-AML使用自适应距离度量学习(AML)算法学习数据的输入空间，该算法是DML算法中的一种。在输入数据空间为高维的情况下，通过优化距离度量来提高SOINN算法的增量聚类性能。在实验结果中，我们使用2个人工数据集、7个来自UCI数据集的真实数据集和3个真实图像数据集来评估性能。我们发现所提出的算法优于传统的SOINN (Shen et al . 2006)和Enhanced SOINN (Shen et al . 2007)算法。与基于SOINN的最先进方法相比，聚类精度(NMI)的改进在0.03到0.13之间。
{"title":"Online incremental clustering with distance metric learning for high dimensional data","authors":"S. Okada, T. Nishida","doi":"10.1109/IJCNN.2011.6033478","DOIUrl":"https://doi.org/10.1109/IJCNN.2011.6033478","url":null,"abstract":"In this paper, we present a novel incremental clustering algorithm which assigns of a set of observations into clusters and learns the distance metric iteratively in an incremental manner. The proposed algorithm SOINN-AML is composed based on the Self-organizing Incremental Neural Network (Shen et al 2006), which represents the distribution of unlabeled data and reports a reasonable number of clusters. SOINN adopts a competitive Hebbian rule for each input signal, and distance between nodes is measured using the Euclidean distance. Such algorithms rely on the distance metric for the input data patterns. Distance Metric Learning (DML) learns a distance metric for the high dimensional input space of data that preserves the distance relation among the training data. DML is not performed for input space of data in SOINN based approaches. SOINN-AML learns input space of data by using the Adaptive Distance Metric Learning (AML) algorithm which is one of the DML algorithms. It improves the incremental clustering performance of the SOINN algorithm by optimizing the distance metric in the case that input data space is high dimensional. In experimental results, we evaluate the performance by using two artificial datasets, seven real datasets from the UCI dataset and three real image datasets. We have found that the proposed algorithm outperforms conventional algorithms including SOINN (Shen et al 2006) and Enhanced SOINN (Shen et al 2007). The improvement of clustering accuracy (NMI) is between 0.03 and 0.13 compared to state of the art SOINN based approaches.","PeriodicalId":415833,"journal":{"name":"The 2011 International Joint Conference on Neural Networks","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129208853","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 : 2011-10-03DOI: 10.1109/IJCNN.2011.6033243
T. Walther, R. Würtz
The problem of learning a generalizable model of the visual appearance of humans from video data is of major importance for computing systems interacting naturally with their users and other humans populating their environment. We propose a step towards automatic behavior understanding by integrating principles of Organic Computing into the posture estimation cycle, thereby relegating the need for human intervention while simultaneously raising the level of system autonomy. The system extracts coherent motion from moving upper bodies and autonomously decides about limbs and their possible spatial relationships. The models from many videos are integrated into meta-models, which show good generalization to different individuals, backgrounds, and attire. These models even allow robust interpretation of single video frames, where all temporal continuity is missing.
{"title":"Autonomous learning of a human body model","authors":"T. Walther, R. Würtz","doi":"10.1109/IJCNN.2011.6033243","DOIUrl":"https://doi.org/10.1109/IJCNN.2011.6033243","url":null,"abstract":"The problem of learning a generalizable model of the visual appearance of humans from video data is of major importance for computing systems interacting naturally with their users and other humans populating their environment. We propose a step towards automatic behavior understanding by integrating principles of Organic Computing into the posture estimation cycle, thereby relegating the need for human intervention while simultaneously raising the level of system autonomy. The system extracts coherent motion from moving upper bodies and autonomously decides about limbs and their possible spatial relationships. The models from many videos are integrated into meta-models, which show good generalization to different individuals, backgrounds, and attire. These models even allow robust interpretation of single video frames, where all temporal continuity is missing.","PeriodicalId":415833,"journal":{"name":"The 2011 International Joint Conference on Neural Networks","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125493444","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 : 2011-10-03DOI: 10.1109/IJCNN.2011.6033568
H. Wagatsuma, Yousuke Tomonaga
The granularity of “action” within a system is highly depending on the internal representation for the task, or intention of what to do if it is a biological system. In the same time, there are several levels of adaptation when the system tries to complete a mission. The problem of choosing the right level of action representation is essential for robot controls as well as in learning paradigms. Both tend to use low-granularity and transfer the processed information to upper levels constructively. However the system never guarantees the completion time of the mission if the system is composed of stiff functional blocks with a specific temporal granularity at the bottom level. However, we biological system have an ability to manage the global time for scheduling and reorganization of tasks to finish by the deadline. Brain-inspired robotics allows us to investigate a distributed parallel information system, the brain, with the ability of time management as a real time control system of the physical body through flexible planning of necessary actions by interacting with the real environment. It is an extension of subsumption approaches that fixed a set of behaviors as the basic unit of action in the viewpoint of temporal property. By focusing on the temporal granularity as a consequence of coordination among multiple levels, a self-confident robot control may arise from a coupling between top-down or purpose-oriented decomposition of the purpose to primitive functions with flexible time windows and bottom-up of sensori-motor reactions in dynamic environments.
{"title":"Problems of temporal granularity in robot control: Levels of adaptation and a necessity of self-confidence","authors":"H. Wagatsuma, Yousuke Tomonaga","doi":"10.1109/IJCNN.2011.6033568","DOIUrl":"https://doi.org/10.1109/IJCNN.2011.6033568","url":null,"abstract":"The granularity of “action” within a system is highly depending on the internal representation for the task, or intention of what to do if it is a biological system. In the same time, there are several levels of adaptation when the system tries to complete a mission. The problem of choosing the right level of action representation is essential for robot controls as well as in learning paradigms. Both tend to use low-granularity and transfer the processed information to upper levels constructively. However the system never guarantees the completion time of the mission if the system is composed of stiff functional blocks with a specific temporal granularity at the bottom level. However, we biological system have an ability to manage the global time for scheduling and reorganization of tasks to finish by the deadline. Brain-inspired robotics allows us to investigate a distributed parallel information system, the brain, with the ability of time management as a real time control system of the physical body through flexible planning of necessary actions by interacting with the real environment. It is an extension of subsumption approaches that fixed a set of behaviors as the basic unit of action in the viewpoint of temporal property. By focusing on the temporal granularity as a consequence of coordination among multiple levels, a self-confident robot control may arise from a coupling between top-down or purpose-oriented decomposition of the purpose to primitive functions with flexible time windows and bottom-up of sensori-motor reactions in dynamic environments.","PeriodicalId":415833,"journal":{"name":"The 2011 International Joint Conference on Neural Networks","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126806438","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 : 2011-10-03DOI: 10.1109/IJCNN.2011.6033319
Zhitang Chen, L. Chan
Permutation indeterminacy and scaling ambiguity occur in ICA and they are particularly problematic in time-frequency domain separation of convolutive mixtures. The quality of separation is severely degraded if these two problems are not well addressed. In this paper, we propose new approaches to solve the permutation indeterminacy and scaling ambiguity in the separation of convolutive mixture in frequency domain. We first apply Short Time Fourier Transform to the observed signals in order to transform the convolutive mixing in time domain to instantaneous mixing in time-frequency domain. A fixed-point algorithm with test of saddle point is adopted to derive the separated components in each frequency bin. To solve the permutation problem,we propose a new matching algorithm for this purpose. First we use discrete Haar Wavelet Transform to extract the feature vectors from the magnitude waveforms of the separated components and use Singular Value Decomposition to achieve dimension reduction. The permutation problem is solved by clustering the feature vectors using the new matching algorithm which is a combination of basic K-means and Hungarian algorithm. To solve the scaling ambiguity problem, we treat it as an overcomplete problem and realize it by maximizing the posterior of the scaling factor. Finally, experiments are conducted using benchmark data to present the effectiveness and performance of our proposed algorithms.
{"title":"New approaches for solving permutation indeterminacy and scaling ambiguity in frequency domain separation of convolved mixtures","authors":"Zhitang Chen, L. Chan","doi":"10.1109/IJCNN.2011.6033319","DOIUrl":"https://doi.org/10.1109/IJCNN.2011.6033319","url":null,"abstract":"Permutation indeterminacy and scaling ambiguity occur in ICA and they are particularly problematic in time-frequency domain separation of convolutive mixtures. The quality of separation is severely degraded if these two problems are not well addressed. In this paper, we propose new approaches to solve the permutation indeterminacy and scaling ambiguity in the separation of convolutive mixture in frequency domain. We first apply Short Time Fourier Transform to the observed signals in order to transform the convolutive mixing in time domain to instantaneous mixing in time-frequency domain. A fixed-point algorithm with test of saddle point is adopted to derive the separated components in each frequency bin. To solve the permutation problem,we propose a new matching algorithm for this purpose. First we use discrete Haar Wavelet Transform to extract the feature vectors from the magnitude waveforms of the separated components and use Singular Value Decomposition to achieve dimension reduction. The permutation problem is solved by clustering the feature vectors using the new matching algorithm which is a combination of basic K-means and Hungarian algorithm. To solve the scaling ambiguity problem, we treat it as an overcomplete problem and realize it by maximizing the posterior of the scaling factor. Finally, experiments are conducted using benchmark data to present the effectiveness and performance of our proposed algorithms.","PeriodicalId":415833,"journal":{"name":"The 2011 International Joint Conference on Neural Networks","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126237003","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 : 2011-10-03DOI: 10.1109/IJCNN.2011.6033228
Rafael Marcos Luque Baena, D. Elizondo, Ezequiel López-Rubio, E. Palomo
In this paper, a novel methodology for using feature selection in hand biometric systems, based on genetic algorithms and mutual information is presented. A hand segmentation algorithm based on adaptive threshold and active contours is also applied, in order to deal with complex backgrounds and non-homogeneous illumination.
{"title":"GA-based feature selection approach in biometric hand systems","authors":"Rafael Marcos Luque Baena, D. Elizondo, Ezequiel López-Rubio, E. Palomo","doi":"10.1109/IJCNN.2011.6033228","DOIUrl":"https://doi.org/10.1109/IJCNN.2011.6033228","url":null,"abstract":"In this paper, a novel methodology for using feature selection in hand biometric systems, based on genetic algorithms and mutual information is presented. A hand segmentation algorithm based on adaptive threshold and active contours is also applied, in order to deal with complex backgrounds and non-homogeneous illumination.","PeriodicalId":415833,"journal":{"name":"The 2011 International Joint Conference on Neural Networks","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122749332","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 : 2011-10-03DOI: 10.1109/IJCNN.2011.6033613
J. Weng
Finite Automata (FA) is a base net for many sophisticated probability-based systems of artificial intelligence. However, an FA processes symbols, instead of images that the brain senses and produces (e.g., sensory images and motor images). Of course, many recurrent artificial neural networks process images. However, their non-calibrated internal states prevent generalization, let alone the feasibility of immediate and error-free learning. I wish to report a general-purpose Developmental Program (DP) for a new type of, brain-anatomy inspired, networks — Developmental Networks (DNs). The new theoretical results here are summarized by three theorems. (1) From any complex FA that demonstrates human knowledge through its sequence of the symbolic inputs-outputs, the DP incrementally develops a corresponding DN through the image codes of the symbolic inputs-outputs of the FA. The DN learning from the FA is incremental, immediate and error-free. (2) After learning the FA, if the DN freezes its learning but runs, it generalizes optimally for infinitely many image inputs and actions based on the embedded inner-product distance, state equivalence, and the principle of maximum likelihood. (3) After learning the FA, if the DN continues to learn and run, it “thinks” optimally in the sense of maximum likelihood based on its past experience.
{"title":"Three theorems: Brain-like networks logically reason and optimally generalize","authors":"J. Weng","doi":"10.1109/IJCNN.2011.6033613","DOIUrl":"https://doi.org/10.1109/IJCNN.2011.6033613","url":null,"abstract":"Finite Automata (FA) is a base net for many sophisticated probability-based systems of artificial intelligence. However, an FA processes symbols, instead of images that the brain senses and produces (e.g., sensory images and motor images). Of course, many recurrent artificial neural networks process images. However, their non-calibrated internal states prevent generalization, let alone the feasibility of immediate and error-free learning. I wish to report a general-purpose Developmental Program (DP) for a new type of, brain-anatomy inspired, networks — Developmental Networks (DNs). The new theoretical results here are summarized by three theorems. (1) From any complex FA that demonstrates human knowledge through its sequence of the symbolic inputs-outputs, the DP incrementally develops a corresponding DN through the image codes of the symbolic inputs-outputs of the FA. The DN learning from the FA is incremental, immediate and error-free. (2) After learning the FA, if the DN freezes its learning but runs, it generalizes optimally for infinitely many image inputs and actions based on the embedded inner-product distance, state equivalence, and the principle of maximum likelihood. (3) After learning the FA, if the DN continues to learn and run, it “thinks” optimally in the sense of maximum likelihood based on its past experience.","PeriodicalId":415833,"journal":{"name":"The 2011 International Joint Conference on Neural Networks","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126374445","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 : 2011-10-03DOI: 10.1109/IJCNN.2011.6033468
J. Wall, T. McGinnity, L. Maguire
This paper outlines the development of a cross-correlation algorithm and a spiking neural network (SNN) for sound localisation based on real sound recorded in a noisy and dynamic environment by a mobile robot. The SNN architecture aims to simulate the sound localisation ability of the mammalian auditory pathways by exploiting the binaural cue of interaural time difference (ITD). The medial superior olive was the inspiration for the SNN architecture which required the integration of an encoding layer which produced biologically realistic spike trains, a model of the bushy cells found in the cochlear nucleus and a supervised learning algorithm. The experimental results demonstrate that biologically inspired sound localisation achieved using a SNN can compare favourably to the more classical technique of cross-correlation.
{"title":"A comparison of sound localisation techniques using cross-correlation and spiking neural networks for mobile robotics","authors":"J. Wall, T. McGinnity, L. Maguire","doi":"10.1109/IJCNN.2011.6033468","DOIUrl":"https://doi.org/10.1109/IJCNN.2011.6033468","url":null,"abstract":"This paper outlines the development of a cross-correlation algorithm and a spiking neural network (SNN) for sound localisation based on real sound recorded in a noisy and dynamic environment by a mobile robot. The SNN architecture aims to simulate the sound localisation ability of the mammalian auditory pathways by exploiting the binaural cue of interaural time difference (ITD). The medial superior olive was the inspiration for the SNN architecture which required the integration of an encoding layer which produced biologically realistic spike trains, a model of the bushy cells found in the cochlear nucleus and a supervised learning algorithm. The experimental results demonstrate that biologically inspired sound localisation achieved using a SNN can compare favourably to the more classical technique of cross-correlation.","PeriodicalId":415833,"journal":{"name":"The 2011 International Joint Conference on Neural Networks","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122240886","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 : 2011-10-03DOI: 10.1109/IJCNN.2011.6033356
D. Anguita, A. Ghio, L. Oneto, S. Ridella
The Structural Risk Minimization framework has been recently proposed as a practical method for model selection in Support Vector Machines (SVMs). The main idea is to effectively measure the complexity of the hypothesis space, as defined by the set of possible classifiers, and to use this quantity as a penalty term for guiding the model selection process. Unfortunately, the conventional SVM formulation defines a hypothesis space centered at the origin, which can cause undesired effects on the selection of the optimal classifier. We propose here a more flexible SVM formulation, which addresses this drawback, and describe a practical method for selecting more effective hypothesis spaces, leading to the improvement of the generalization ability of the final classifier.
{"title":"Selecting the hypothesis space for improving the generalization ability of Support Vector Machines","authors":"D. Anguita, A. Ghio, L. Oneto, S. Ridella","doi":"10.1109/IJCNN.2011.6033356","DOIUrl":"https://doi.org/10.1109/IJCNN.2011.6033356","url":null,"abstract":"The Structural Risk Minimization framework has been recently proposed as a practical method for model selection in Support Vector Machines (SVMs). The main idea is to effectively measure the complexity of the hypothesis space, as defined by the set of possible classifiers, and to use this quantity as a penalty term for guiding the model selection process. Unfortunately, the conventional SVM formulation defines a hypothesis space centered at the origin, which can cause undesired effects on the selection of the optimal classifier. We propose here a more flexible SVM formulation, which addresses this drawback, and describe a practical method for selecting more effective hypothesis spaces, leading to the improvement of the generalization ability of the final classifier.","PeriodicalId":415833,"journal":{"name":"The 2011 International Joint Conference on Neural Networks","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126056257","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 : 2011-10-03DOI: 10.1109/IJCNN.2011.6033469
F. Mokhayeri, M. Akbarzadeh-T.
This paper presents a facial expression recognition approach to recognize the affective states. Feature extraction is a vital step in the recognition of facial expressions. In this work, a novel facial feature extraction method based on Intersecting Cortical Model (ICM) is proposed. The ICM network which is a simplified model of Pulse-Coupled Neural Network (PCNN) model has great potential to perform pixel grouping. In the proposed method the normalized face image is segmented into two regions including mouth, eyes using fuzzy c-means clustering (FCM). Segmented face images are imported into an ICM network with 300 iteration number and pulse image produced by the ICM network is chosen as the face code, then the support vector machine (SVM) is trained for discrimination of different expressions to distinguish the different affective states. In order to evaluate the performance of the proposed algorithm, the face image dataset is constructed and the proposed algorithm is used to classify seven basic expressions including happiness, sadness, fear, anger, surprise and hate The experimental results confirm that ICM network has great potential for facial feature extraction and the proposed method for human affective recognition is promising. Fast feature extraction is the most advantage of this method which can be useful for real world application.
{"title":"A novel facial feature extraction method based on ICM network for affective recognition","authors":"F. Mokhayeri, M. Akbarzadeh-T.","doi":"10.1109/IJCNN.2011.6033469","DOIUrl":"https://doi.org/10.1109/IJCNN.2011.6033469","url":null,"abstract":"This paper presents a facial expression recognition approach to recognize the affective states. Feature extraction is a vital step in the recognition of facial expressions. In this work, a novel facial feature extraction method based on Intersecting Cortical Model (ICM) is proposed. The ICM network which is a simplified model of Pulse-Coupled Neural Network (PCNN) model has great potential to perform pixel grouping. In the proposed method the normalized face image is segmented into two regions including mouth, eyes using fuzzy c-means clustering (FCM). Segmented face images are imported into an ICM network with 300 iteration number and pulse image produced by the ICM network is chosen as the face code, then the support vector machine (SVM) is trained for discrimination of different expressions to distinguish the different affective states. In order to evaluate the performance of the proposed algorithm, the face image dataset is constructed and the proposed algorithm is used to classify seven basic expressions including happiness, sadness, fear, anger, surprise and hate The experimental results confirm that ICM network has great potential for facial feature extraction and the proposed method for human affective recognition is promising. Fast feature extraction is the most advantage of this method which can be useful for real world application.","PeriodicalId":415833,"journal":{"name":"The 2011 International Joint Conference on Neural Networks","volume":"34 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125615036","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：