Search

Contains fulltext : 81430.pdf (author's version ) (Closed access)

This work proposes a concept for indoor ambulatory monitoring for Parkinson's disease patients. In the proposed concept, a wearable inertial sensor is kept as the main monitoring device through the day, and it is expanded by an ambient sensor system in the specific living areas with high estimated probability of occurrence of freezing of gait episode. The ambient sensor system supports decisions of the wearable sensor system by providing relevant spatial context information of the user, which is obtained through precise localization.

We extend the self-organizing map to a supervised fuzzy classification method. On the one hand, this leads to a robust classifier where efficient learning of fuzzy labeled or partially contradictory data is possible. On the other hand, class similarities may be detected. Further, we integrate the parametrized functional metric as data similarity measure into the approach. Parametrization of this functional metric allows relevance learning for efficient classification and feature selection.

Prototype based neural clustering or data mining methods such as the self-organizing map or neural gas constitute intuitive and powerful machine learning tools for a variety of application areas. However, the classical methods are restricted to data embedded in a real vector space and have only limited applicability to noneuclidean data as occurs in, for example, biomedical or symbolic fields. Recently, extensions of unsupervised neural prototype based clustering to dissimilarity data, i.e. data characterized in terms of a dissimilarity matrix only, have been proposed substituting the mean by the so-called generalized median. Thereby, the location of prototypes is chosen within the discrete input space which constitutes a severe limitation in particular for sparse data sets since the prototype flexibility is restricted. Here we present a generalization of median neural gas such that prototypes can be interpreted as mixtures of discrete input locations. We derive a batch optimization scheme based on a corresponding cost function.

Neural networks are intended to be used in future nanoelectronic technology since these architectures seem to be robust to malfunctioning elements and noise in its inputs and parameters. In this work, the robustness of radial basis function networks is analyzed in order to operate in noisy and unreliable environment. Furthermore, upper bounds on the mean square error under noise contaminated parameters and inputs are determined if the network parameters are constrained. To achieve robuster neural network architectures fundamental methods are introduced to identify sensitive parameters and neurons.

In the field of explorative data analysis self-organizing maps have been used successfully for a lot of applications. In our case, we apply the self-organizing map for the analysis of semiconductor fabrication data by training recorded high dimensional data sets. Usually, the training result is displayed by using appropriate visualization techniques and the results are evaluated manually. Especially for large data sets an automated post-processing of the training result is essential. In this paper an automatic training result analysis based on specific image processing is introduced. Dependencies between components maps are calculated by structure overlapping analysis based on the segmentation of component maps. This novel method has been integrated into the data analysis software DanI, that simulates self-organizing maps for data analysis with several pre-processing and post-processing capabilities.

(http://www.mpae.gwdg.de/EGS/egsga/denhaag99/prg.htm); International audience

International audience; The sensitivity of a circuit dedicated to fuzzy control (the WARP 2.0 from SGS-Thomson) is investigated with respect to "single event upsets" caused by nuclear or space environments. The fuzzy control part of a future ESA Mars instrument deployment vehicle serves as a realistic test application. Software fault injection experiments and ground tests show the fault tolerance properties of the studied application.

International audience; Two different approaches, used for the analysis of textures within SPOT images, are compared from both the performances and the robustness points of views. Fault injection simulations performed on a digital prototype show that the two approaches are suitable for on-board satellite image processing

The training of a neural network is an intricate balance between knowledge, randomness and symmetry. Symmetry can both be beneficial and detrimental to the learning process by respectively equality of choice and indecision. The paper provides a critical review and classification, and offers a constructive procedure to handle problem-indigenous symmetries.

Neural networks can be applied for a number of innovative applications in a production environment, ranging from security & safety in the environmental conditions to the product control & diagnosis. For visual monitoring the use of low-resolution images is promising to bridge the time elapse between more elaborous stampings. This facility for biologically motivated decision-making is illustrated by two real-life applications: stain detection and mixture diagnosis.