U
Ulrich Rückert
Researcher at Bielefeld University
Publications - 363
Citations - 3446
Ulrich Rückert is an academic researcher from Bielefeld University. The author has contributed to research in topics: Robot & Field-programmable gate array. The author has an hindex of 26, co-authored 357 publications receiving 3235 citations. Previous affiliations of Ulrich Rückert include Technische Universität München & Technical University of Dortmund.
Papers
More filters
Book ChapterDOI
Tolerance of radial basis functions against stuck-at-faults
Ralf Eickhoff,Ulrich Rückert +1 more
TL;DR: This paper analyzes the fault-tolerance of Radial Basis Function networks to Stuck-At-Faults at the trained weights and at the output of neurons and determines upper bounds on the mean square error arising from these faults.
Book ChapterDOI
VLSI Design of an Associative Memory Based on Distributed Storage of Information
TL;DR: Artificial neural networks (ANNs) are well suited for the implementation of associative memories at least because the processing elements (artificial neurons) in an ANN operate in a highly parallel way and thus a considerable gain in speed is to be expected.
Proceedings Article
Jointly Trained Variational Autoencoder for Multi-Modal Sensor Fusion
TL;DR: This work presents the novel multi-modal Variational Autoencoder approach which is derived from the complete marginal joint log-likelihood, which allows the end-to-end training of Bayesian information fusion on raw data for all subsets of a sensor setup.
Proceedings ArticleDOI
Placement-oriented modeling of partially reconfigurable architectures
TL;DR: A new analytical model is developed which can be used for the analysis of the various approaches to dynamic reconfiguration and is able to consider miscellaneous constraints and cost parameters for online placement.
A framework for design space exploration of resource efficient network processing on multiprocessor SoCs
TL;DR: In this paper, the authors present a multiprocessor architecture that employs a network-on-chip (NoC) to achieve a high degree of scalability for the application and for the system in respect to future integration technologies.