Towards Understanding the Role of Regulatory Feedback in Simultaneous Pattern Processing
-
1
Los Alamos National Labs, T5 Theoretical Division, United States
Simultaneous Pattern Processing (SPP) is the ability to identify simultaneous-intermixed patterns without isolating them individually (e.g. without separating each pattern in space and processing it individually). This ability appears to separate human from artificial algorithm capabilities.
Enhanced SPP ability is beneficial for many real-life applications such as scene understanding, separating simultaneous voices “cocktail party problem”, and identifying odorant or taste mixes.
This work identifies the type of SPP scenarios that are problematic to artificial algorithms and quantifies how subtle difficulties lead to complex combinatorial issues. In SPP Test patterns are a superposition of multiple trained patterns. If random single patterns are trained labeled A..Z then simultaneous patterns are composed of combinations: A+B, A+C ... or A+B+C etc. All possible combinations of up to 8 simultaneously are evaluated.
The second part of this work proposes an algorithm motivated by ubiquitous re-entrant “output to input” connections found throughout sensory processing regions of the brain. This architecture regulates the amount of input activity available to outputs (dynamically determining salience) and is demonstrated to have enhanced SPP ability. Thus, the brain may utilize this configuration for enhanced SPP performance.
Keywords:
computational neuroscience
Conference:
Bernstein Conference on Computational Neuroscience, Berlin, Germany, 27 Sep - 1 Oct, 2010.
Presentation Type:
Presentation
Topic:
Bernstein Conference on Computational Neuroscience
Citation:
Achler
T
(2010). Towards Understanding the Role of Regulatory Feedback in Simultaneous Pattern Processing.
Front. Comput. Neurosci.
Conference Abstract:
Bernstein Conference on Computational Neuroscience.
doi: 10.3389/conf.fncom.2010.51.00035
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
17 Sep 2010;
Published Online:
23 Sep 2010.
*
Correspondence:
Dr. Tsvi Achler, Los Alamos National Labs, T5 Theoretical Division, Los Alamos, United States, achler@gmail.com