Echo State Networks in Dynamic Data Clustering

Koprinkova–Hristova, Petia; Alexiev, Kiril

doi:10.1007/978-3-642-40728-4_43

Cited by 12 publications

(14 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the present paper we move one step further applying a two dimensional density distribution functions and using a procedure from [12] that determines automatically the kernels bandwidth. The improved algorithm was applied for clustering of spectral satellite images -a data set with much more dimensions than those used in our initial work.…”

Section: Introductionmentioning

confidence: 99%

“…These extracted features were used further as inputs to a feedforward neural network classifier. In [12] we exploited the same reservoir ability but looking from another perspective: we consider combinations between steady states of each two neurons in the reservoir as numerous twodimensional projections of the original multidimensional data fed into ESN input. These low dimensional projections can be used next for data clustering.…”

Section: Introductionmentioning

confidence: 99%

“…The initial investigations in [12] used two small size data sets. But even with these data need of further refinement of our procedure with respect to finding of a proper two dimensional projection among the numerous possible combinations between reservoir neurons was revealed.…”

Section: Introductionmentioning

confidence: 99%

“…It is shown that with new improvement we are able to discriminate easier one among all possible two dimensional projections. Figure 1 below represents the basic structure of an ESN as it was used in [12]. The ESN reservoir dynamics is described as follows [6]:…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Clustering of spectral images using Echo state networks

Koprinkova–Hristova¹,

Angelova²,

Borisova

et al. 2013

2013 Ieee Inista

Self Cite

View full text Add to dashboard Cite

In the present work we applied a recently developed procedure for multidimensional data clustering to processing of spectral satellite images. The core of our approach lays in projection of multidimensional image to a two dimensional one. The main aim is to discover points with similar characteristics. This was done by clustering of the resulting image. The processing technique exploits equilibrium states of a kind of recurrent neural network -Echo state network (ESN) -that are obtained after intrinsic plasticity (IP) tuning of the ESN using multidimensional data as inputs. The proposed in our previous work automated procedure for multidimensional data clustering is further refined and tested on the satellite image data. The obtained number and position of clusters of a multispectral image of a mountain region in Bulgaria is compared with the classification of the region landscape given by the Ministry of Regional Development and Public Works.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Clustering of spectral images using Echo state networks

Koprinkova–Hristova¹,

Angelova²,

Borisova

et al. 2013

2013 Ieee Inista

Self Cite

View full text Add to dashboard Cite

show abstract

“…Artificial neural network (ANN), also known as simulated neural network, refers to an interconnected artificial neuron group that uses mathematical or computational models for information processing [1]. The learning process used to train artificial neural networks is itself a statistical technique, and White H [2] proposed some potentially useful new training methods for artificial neural networks.…”

Section: Introductionmentioning

confidence: 99%

Application research of image recognition technology based on CNN in image location of environmental monitoring UAV

Zhao

He²,

Wu³

et al. 2018

J Image Video Proc.

View full text Add to dashboard Cite

UAV remote sensing has been widely used in emergency rescue, disaster relief, environmental monitoring, urban planning, and so on. Image recognition and image location in environmental monitoring has become an academic hotspot in the field of computer vision. Convolution neural network model is the most commonly used image processing model. Compared with the traditional artificial neural network model, convolution neural network has more hidden layers. Its unique convolution and pooling operations have higher efficiency in image processing. It has incomparable advantages in image recognition and location and other forms of two-dimensional graphics tasks. As a new deformation of convolution neural network, residual neural network aims to make convolution layer learn a kind of residual instead of a direct learning goal. After analyzing the characteristics of CNN model for image feature representation and residual network, a residual network model is built. The UAV remote sensing system is selected as the platform to acquire image data, and the problem of image recognition based on residual neural network is studied, which is verified by experiment simulation and precision analysis. Finally, the problems and experiences in the process of learning and designing are discussed, and the future improvements in the field of image target location and recognition are prospected.

show abstract

Time Series Clustering with Deep Reservoir Computing

Atencia

Gallicchio

Joya

et al. 2020

Artificial Neural Networks and Machine Learning – ICANN 2020

View full text Add to dashboard Cite

This paper proposes a method for clustering of time series, based upon the ability of deep Reservoir Computing networks to grasp the dynamical structure of the series that is presented as input. A standard clustering algorithm, such as k-means, is applied to the network states, rather than the input series themselves. Clustering is thus embedded into the network dynamical evolution, since a clustering result is obtained at every time step, which in turn serves as initialisation at the next step. We empirically assess the performance of deep reservoir systems in time series clustering on benchmark datasets, considering the influence of crucial hyper-parameters. Experimentation with the proposed model shows enhanced clustering quality, measured by the silhouette coefficient, when compared to both static clustering of data, and dynamic clustering with a shallow network.

show abstract

Echo State Networks in Dynamic Data Clustering

Cited by 12 publications

References 13 publications

Clustering of spectral images using Echo state networks

Clustering of spectral images using Echo state networks

Application research of image recognition technology based on CNN in image location of environmental monitoring UAV

Time Series Clustering with Deep Reservoir Computing

Contact Info

Product

Resources

About