Inspired by the recent successes of boosting algorithms, a trend in unsupervised learning has begun to emphasize the need to explore the design of weighted clustering algorithms. In this paper we handle clustering as ...
详细信息
ISBN:
(纸本)0769521282
Inspired by the recent successes of boosting algorithms, a trend in unsupervised learning has begun to emphasize the need to explore the design of weighted clustering algorithms. In this paper we handle clustering as a constrained minimization of a Bregman divergence. theoretical results show benefits resembling those of boosting algorithms, and bring new modified weighted versions of clustering algorithms such as k-means, expectation-maximization (EM) and k-harmonic means. Experiments display the quality of the results obtained, and corroborate the advantages that subtle data reweightings may indeed bring to clustering.
In this paper we propose an Isomap-based nonlinear alternative to the linear subspace method for manifold representation of view-varying faces. Being interested in user-independent head pose estimation, we extend the ...
详细信息
ISBN:
(纸本)0769521282
In this paper we propose an Isomap-based nonlinear alternative to the linear subspace method for manifold representation of view-varying faces. Being interested in user-independent head pose estimation, we extend the Isomap model [1] to be able to map (high-dimensional) input data points which are not in the training data set into the dimensionality-reduced space found by the model From this representation, a pose parameter map relating the input face samples to view angles is learnt. the proposed method is evaluated on a large database of multi-view face images in comparison to two other recently proposed subspace methods.
We propose an original learning approach for image classification problems. Recognizing semantic events in video requires to preliminary learn the different classes of events. this first stage is crucial since it cond...
详细信息
ISBN:
(纸本)0769521282
We propose an original learning approach for image classification problems. Recognizing semantic events in video requires to preliminary learn the different classes of events. this first stage is crucial since it conditions the further classification results. In video content analysis, the task is especially difficult due to the high intra-class variability and to noisy measurements. We then represent each class by the centers of several sub-classes (or clusters) thanks to a robust partitional clustering algorithm which can be applied in parallel to a (non-predefined) number of classes. Our clustering technique overcome three main limitations of standard K-means methods: sensitivity to initialization, choice of the number of clusters and influence of outliers. Moreover, it can process the training data in an incremental way. Experimental results on sports videos are reported.
the nearest neighbour (NN) classification rule is usually chosen in a large number of patternrecognition systems due to its simplicity and good properties. As the problem of finding the nearest neighbour of an unknow...
详细信息
ISBN:
(纸本)0769521282
the nearest neighbour (NN) classification rule is usually chosen in a large number of patternrecognition systems due to its simplicity and good properties. As the problem of finding the nearest neighbour of an unknown sample is also of interest in other scientific communities (very large databases, datamining, computational geometry, ...), a vast number of fast nearest neighbour search algorithms have been developed during the last years. In order to improve classification rates, the k-NN rule is often used instead of the NN rule, but it yields higher classification times. In this work we introduce a new classification rule applicable to many of those algorithms in order to obtain classification rates better than those of the nearest neighbour (similar to those of the k-NN rule) without significantly increasing classification time.
Self-organizing neural networks achieve more predictable and accurate results then the classic ones withthe static architecture. Neurons and connections of such neural networks are dynamically built during the learni...
详细信息
ISBN:
(纸本)0769521282
Self-organizing neural networks achieve more predictable and accurate results then the classic ones withthe static architecture. Neurons and connections of such neural networks are dynamically built during the learning process. Self-organizing neural networks based on the Group Method of data Handling (GMDH) have proven to be one of the most efficient approaches to solving the problems of patternrecognition withthe statistical learningdata. In this article we propose a new method for searching deeper interrelations of the inputs and the output of the system under the study of such a neural network. the method allows eliminating links to the inputs that are no longer useful at the later steps of the neural network construction, thus allowing to simplify the neural network structure and increase prediction accuracy. Hence the method is called the Structure Relaxation Method. For complex problems the method helps to find deeper system inputs interrelations, increase the prediction accuracy, and, at the same time, decrease the number of the inputs being used. the proposed relaxation method was tested on the real world problems;the results are also presented herein.
K-Means clustering is a well-known partition-based technique in unsupervised learning to construct pattern models. the main difficulty, however, is that its performance is highly susceptible to the initialized partiti...
详细信息
ISBN:
(纸本)0769521282
K-Means clustering is a well-known partition-based technique in unsupervised learning to construct pattern models. the main difficulty, however, is that its performance is highly susceptible to the initialized partition. To attack this problem, a suboptimal K-Means algorithm is briefly reviewed by applying dynamic programming over the principal component direction. In particular, a heuristic clustering dissimilarity, the Delta-MSE function, is incorporated into the suboptimal K-Means algorithm. the Delta-MSE function is derived by calculating the difference of within-class variance before and after moving a given data sample from one cluster to another. Experimental results show that the suboptimal K-Means algorithm that uses the Delta-MSE dissimilarity generally outperforms the original L-2 distance based suboptimal algorithm and a specific kd-tree clustering algorithm.
In this paper we present the application of machinelearning techniques to the improvement of the congestion control of TCP in wired/wireless networks. TCP is suboptimal in hybrid wired/wireless networks because it re...
详细信息
ISBN:
(纸本)0769521428
In this paper we present the application of machinelearning techniques to the improvement of the congestion control of TCP in wired/wireless networks. TCP is suboptimal in hybrid wired/wireless networks because it reacts in the same way to losses due to congestion and losses due to link errors. We thus propose to use machinelearning techniques to build automatically a loss classifier from a database obtained by simulations of random network topologies. Several machinelearning algorithms are compared for this task and the best method for this application turns out to be decision tree boosting. It outperforms ad hoc classifiers proposed in the networking literature.
We present a novel representation of cyclic human locomotion based on a set of spatio-temporal curves of tracked points on the surface of a person. We start by extracting a set of continuous, phase aligned spatio-temp...
详细信息
ISBN:
(纸本)0769521282
We present a novel representation of cyclic human locomotion based on a set of spatio-temporal curves of tracked points on the surface of a person. We start by extracting a set of continuous, phase aligned spatio-temporal curves from trajectories of random points tracked over several cycles of locomotion in a monocular video sequence. We analyze a PCA representation of a set of cyclic curves, pointing out properties of the representation which can be used for spatio-temporal alignment in tracking and recognition tasks. We model the curve distribution density by a mixture of Gaussians using expectation-maximization algorithm. For recognition, we use maximum a posteriori estimate combined with linear data adaptation. We tested the algorithms on CMU MoBo database with favourable results for the recognition of people "by walking" from monocular video sequences captured from the side view.
the pattern model representation (PMR) of time series is proposed in this paper. PMR is based on a piecewise linear representation (PLR) and is effective at describing the tendency of time series. then, the pattern di...
详细信息
ISBN:
(纸本)1853128066
the pattern model representation (PMR) of time series is proposed in this paper. PMR is based on a piecewise linear representation (PLR) and is effective at describing the tendency of time series. then, the pattern distance can be calculated to measure the similarity of tendency. this method overcomes the problem of time series mismatch based on point distance. According to the numbers of series' segmentations, pattern distance has a multi-scale feature and can reflect different similarities with various bandwidths. Because normalization is unnecessary, the calculation consumption of pattern distance is low.
We propose a novel query-driven lazy teaming algorithm which attempts to discover useful local patterns, called support patterns, for classifying a given query. the teaming is customized to the query to avoid the hori...
详细信息
ISBN:
(纸本)0769521428
We propose a novel query-driven lazy teaming algorithm which attempts to discover useful local patterns, called support patterns, for classifying a given query. the teaming is customized to the query to avoid the horizon effect. We show that this query-driven teaming algorithm can guarantee to discover all support patterns with perfect expected accuracy in polynomial time. the experimental results on benchmark data sets also demonstrate that our teaming algorithm really has prominent learning performance.
暂无评论