Header

UZH-Logo

Maintenance Infos

A Framework of Relational Networks to Build Systems with Sensors able to Perform the Joint Approximate Inference of Quantities


Martel, Julien; Cook, Matthew (2015). A Framework of Relational Networks to Build Systems with Sensors able to Perform the Joint Approximate Inference of Quantities. In: IROS 2015, Workshop on Advances in Biologically Inspired Brain-Like Cognition and Control for Learning Robots, Hamburg, 1 October 2015 - 2 October 2015.

Abstract

Probabilistic approaches such as Bayesian inference have been extensively used to design systems able to operate in environments under uncertainty. Implementing these approaches on real-world systems constrained in their latencies or in their power-budget is a challenge because of the general computational intensity required by such methods. In this work, we propose a very simple yet efficient framework to perform approximate inference in a network of quantities between which relations are specified a-priori. We present how we can take advantage of computational features of our framework to implement it in dedicated hardware devices such as GPGPUs or Cellular Processor Arrays (CPAs) for which we demonstrate a simple vision system instantiating the principles of our approach.

Abstract

Probabilistic approaches such as Bayesian inference have been extensively used to design systems able to operate in environments under uncertainty. Implementing these approaches on real-world systems constrained in their latencies or in their power-budget is a challenge because of the general computational intensity required by such methods. In this work, we propose a very simple yet efficient framework to perform approximate inference in a network of quantities between which relations are specified a-priori. We present how we can take advantage of computational features of our framework to implement it in dedicated hardware devices such as GPGPUs or Cellular Processor Arrays (CPAs) for which we demonstrate a simple vision system instantiating the principles of our approach.

Statistics

Downloads

56 downloads since deposited on 04 Feb 2016
27 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Conference or Workshop Item (Speech), refereed, original work
Communities & Collections:07 Faculty of Science > Institute of Neuroinformatics
Dewey Decimal Classification:570 Life sciences; biology
Language:English
Event End Date:2 October 2015
Deposited On:04 Feb 2016 09:45
Last Modified:05 Jul 2016 13:18
Publisher:Proceedings of the IEEE/RSJ international conference on intelligent robots and systems 2015
Series Name:IEEE/RSJ International Conference on Intelligent Robots and Systems, Workshop on Unconventional Computing for Bayesian Inference, IROS 2015

Download

Preview Icon on Download
Preview
Content: Accepted Version
Language: English
Filetype: PDF
Size: 388kB

TrendTerms

TrendTerms displays relevant terms of the abstract of this publication and related documents on a map. The terms and their relations were extracted from ZORA using word statistics. Their timelines are taken from ZORA as well. The bubble size of a term is proportional to the number of documents where the term occurs. Red, orange, yellow and green colors are used for terms that occur in the current document; red indicates high interlinkedness of a term with other terms, orange, yellow and green decreasing interlinkedness. Blue is used for terms that have a relation with the terms in this document, but occur in other documents.
You can navigate and zoom the map. Mouse-hovering a term displays its timeline, clicking it yields the associated documents.

Author Collaborations